IFIP AICT 691

Erlend Alfnes
Anita Romsdal
Jan Ola Strandhagen
Gregor von Cieminski
David Romero (Eds.)

Advances in Production
Management Systems
Production Management Systems
for Responsible Manufacturing,
Service, and Logistics Futures

IFIP WG 5.7 International Conference, APMS 2023

Trondheim, Norway, September 17–21, 2023
Proceedings, Part III

123
IFIP Advances in Information
and Communication Technology 691

Editor-in-Chief

Kai Rannenberg, Goethe University Frankfurt, Germany

Editorial Board Members

TC 1 – Foundations of Computer Science
Luís Soares Barbosa , University of Minho, Braga, Portugal
TC 2 – Software: Theory and Practice
Michael Goedicke, University of Duisburg-Essen, Germany
TC 3 – Education
Arthur Tatnall , Victoria University, Melbourne, Australia
TC 5 – Information Technology Applications
Erich J. Neuhold, University of Vienna, Austria
TC 6 – Communication Systems
Burkhard Stiller, University of Zurich, Zürich, Switzerland
TC 7 – System Modeling and Optimization
Lukasz Stettner, Institute of Mathematics, Polish Academy of Sciences, Warsaw, Poland
TC 8 – Information Systems
Jan Pries-Heje, Roskilde University, Denmark
TC 9 – ICT and Society
David Kreps , National University of Ireland, Galway, Ireland
TC 10 – Computer Systems Technology
Achim Rettberg, Hamm-Lippstadt University of Applied Sciences, Hamm, Germany
TC 11 – Security and Privacy Protection in Information Processing Systems
Steven Furnell , Plymouth University, UK
TC 12 – Artificial Intelligence
Eunika Mercier-Laurent , University of Reims Champagne-Ardenne, Reims, France
TC 13 – Human-Computer Interaction
Marco Winckler , University of Nice Sophia Antipolis, France
TC 14 – Entertainment Computing
Rainer Malaka, University of Bremen, Germany
IFIP Advances in Information and Communication Technology
The IFIP AICT series publishes state-of-the-art results in the sciences and technologies of
information and communication. The scope of the series includes: foundations of
computer science; software theory and practice; education; computer applications in
technology; communication systems; systems modeling and optimization; information
systems; ICT and society; computer systems technology; security and protection in
information processing systems; artificial intelligence; and human-computer interaction.
Edited volumes and proceedings of refereed international conferences in computer
science and interdisciplinary fields are featured. These results often precede journal
publication and represent the most current research.
The principal aim of the IFIP AICT series is to encourage education and the
dissemination and exchange of information about all aspects of computing.

More information about this series at https://link.springer.com/bookseries/6102

Erlend Alfnes Anita Romsdal
• •

Jan Ola Strandhagen Gregor von Cieminski

• •

David Romero
Editors

Advances in Production
Management Systems
Production Management Systems
for Responsible Manufacturing,
Service, and Logistics Futures
IFIP WG 5.7 International Conference, APMS 2023
Trondheim, Norway, September 17–21, 2023
Proceedings, Part III

123
Editors
Erlend Alfnes Anita Romsdal
Norwegian University of Science Norwegian University of Science
and Technology and Technology
Trondheim, Norway Trondheim, Norway
Jan Ola Strandhagen Gregor von Cieminski
Norwegian University of Science ZF Friedrichshafen AG
and Technology Friedrichshafen, Germany
Trondheim, Norway
David Romero
Tecnológico de Monterrey
Mexico City, Mexico

ISSN 1868-4238 ISSN 1868-422X (electronic)

IFIP Advances in Information and Communication Technology
ISBN 978-3-031-43669-7 ISBN 978-3-031-43670-3 (eBook)
https://doi.org/10.1007/978-3-031-43670-3
© IFIP International Federation for Information Processing 2023

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the
material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now
known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book are
believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors
give a warranty, expressed or implied, with respect to the material contained herein or for any errors or
omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Paper in this product is recyclable.

 Preface

The year 2023 has undoubtedly been a year of contrasts. We are experiencing stunning
developments in technology, and creating new products, services, and systems that are
changing the way we live and work. Simultaneously, we are experiencing multiple
conflicts around the world and the brutal effects of climate change. While many
experience success and improved standards of living, others face threats to their lives
and even loss. A Scientific Conference cannot change this but can be seen as a symbol
for aiming for a different future. We create new knowledge and solutions, we share all
our achievements, and we meet to create new friendships and meet people from all over
the world.
The International Conference on “Advances in Production Management Systems”
(APMS) 2023 is the leading annual event of the IFIP Working Group (WG) 5.7 of the
same name. At the Conference in Trondheim, Norway, hosted by the Norwegian
University of Science and Technology (NTNU), more than 200 papers were presented
and discussed. This is a significant step up from the first APMS Conference in 1980,
which assembled just a few participants. The IFIP WG5.7 was established in 1978 by
the General Assembly of the International Federation for Information Processing (IFIP)
in Oslo, Norway. Its first meeting was held in August 1979 with all its seven members
present. The WG has since grown to 108 full members and 25 honorary members.
After 43 years, APMS has returned to the city where it started. The venue in 1980
was Lerchendal Gård, and the topic marked the turn of a decade: “Production Planning
and Control in the 80s”. The papers presented attempted to look into the future – a
future which at that time was believed to be fully digitalized. One foresaw that during
the coming decade, full automation and optimization of complete manufacturing plants,
controlled by a central computer, would be a reality. The batch processing of pro-
duction plans would be replaced by online planning and control systems.
No other technology can show a more rapid development and impact in industry and
society than Information and Communication Technology (ICT). The APMS 2023
program shows that the IFIP WG5.7 still can make and will continue to make a
significant contribution to production and production management disciplines.
In 2023, the International Scientific Committee for APMS included 215 recognized
experts working in the disciplines of production and production management systems.
For each paper, an average of 2.5 single-blind reviews were provided. Over two
months, each submitted paper went through two rigorous rounds of reviews to allow
authors to revise their work after the first round of reviews to guarantee the highest
scientific quality of the papers accepted for publication. Following this process, 213 full
papers were selected for inclusion in the conference proceedings from a total of 224
submissions.
APMS 2023 brought together leading international experts from academia, industry,
and government in the areas of production and production management systems to
discuss how to achieve responsible manufacturing, service, and logistics futures. This
vi Preface

included topics such as innovative manufacturing, service, and logistics systems

characterized by their agility, circularity, digitalization, flexibility, human-centricity,
resiliency, and smartification contributing to more sustainable industrial futures that
ensure that products and services are manufactured, servitized, and distributed in a way
that creates a positive effect on the triple bottom line.
The APMS 2023 conference proceedings are organized into four volumes, covering
a large spectrum of research addressing the overall topic of the conference “Production
Management Systems for Responsible Manufacturing, Service, and Logistics Futures”.
We would like to thank all contributing authors for their quality research work and
their willingness to share their findings with the APMS and IFIP WG5.7 community.
We are equally grateful for the outstanding work of all the International Reviewers, the
Program Committee Members, and the Special Sessions Organizers.

September 2023 Erlend Alfnes

Anita Romsdal
Jan Ola Strandhagen
Gregor von Cieminski
David Romero
 Organization

Conference Chair
Jan Ola Strandhagen Norwegian University of Science and Technology,
Norway

Conference Co-chair
Gregor von Cieminski ZF Friedrichshafen AG, Germany

Conference Honorary Chair

Asbjørn Rolstadås Norwegian University of Science and Technology,
Norway

Program Chair
Erlend Alfnes Norwegian University of Science and Technology,
Norway

Program Co-chairs
Heidi Carin Dreyer Norwegian University of Science and Technology,
Norway
Daryl Powell Norwegian University of Science and Technology/
SINTEF Manufacturing, Norway
Bella Nujen Norwegian University of Science and Technology,
Norway
Anita Romsdal Norwegian University of Science and Technology,
Norway
David Romero Tecnológico de Monterrey, Mexico

Organization Committee Chair

Anita Romsdal Norwegian University of Science and Technology,
Norway

Doctoral Workshop Chair

Hans-Henrik Hvolby Aalborg University, Denmark
viii Organization

Doctoral Workshop Co-chair

David Romero Tecnológico de Monterrey, Mexico

List of Reviewers

Federica Acerbi Ayoub Chakroun

Luca Adelfio Zuhara Chavez
Natalie Cecilia Agerskans Ferdinando Chiacchio
El-Houssaine Aghezzaf Steve Childe
Rajeev Agrawal Chiara Cimini
Carla Susana Agudelo Assuad Florian Clemens
Kosmas Alexopoulos Beatrice Colombo
Kartika Nur Alfina Federica Costa
Erlend Alfnes Catherine da Cunha
Antonio Pedro Dias Alves de Campos Flávia de Souza
Terje Andersen Yüksel Değirmencioğlu Demiralay
Joakim Andersson Enes Demiralay
Dimitris Apostolou Tabea Marie Demke
Germán Arana Landín Mélanie Despeisse
Simone Arena Candice Destouet
Emrah Arica Slavko Dolinsek
Veronica Arioli Milos Drobnjakovic
Nestor Fabián Ayala Eduardo e Oliveira
Christiane Lima Barbosa Malin Elvin
Mohadese Basirati Christos Emmanouilidis
Mohamed Ben Ahmed Hakan Erdeş
Justus Aaron Benning Kristian Johan Ingvar Ericsson
Aili Biriita Bertnum Victor Eriksson
Belgacem Bettayeb Adrodegari Federico
Seyoum Eshetu Birkie Matteo Ferrazzi
Umit Sezer Bititci Jannick Fiedler
Klas Boivie Erik Flores-García
Alexandros Bousdekis Giuseppe Fragapane
Nadjib Brahimi Chiara Franciosi
Greta Braun Susanne Franke
Gianmarco Bressanelli Enzo Frazzon
Jim J. Browne Stefano Frecassetti
Patrick Bründl Jan Frick
Kay Burow Paolo Gaiardelli
Jenny Bäckstrand Clarissa A. González Chávez
Jannicke Baalsrud Hauge Jon Gosling
Robisom Damasceno Calado Danijela Gračanin
Luis Manuel Camarinha-Matos Daniela Greven
Violetta Giada Cannas Eric Grosse
 Organization ix

Zengxu Guo Mohamed Naim

Christopher Gustafsson Farah Naz
Petter Haglund Torbjørn Netland
Lise Lillebrygfjeld Halse Phu Nguyen
Trond Halvorsen Kjeld Nielsen
Robin Hanson Ana Nikolov
Stefanie Hatzl Sang Do Noh
Theresa-Franziska Hinrichsen Antonio Padovano
Maria Holgado Julia Pahl
Christian Holper Martin Perau
Djerdj Horvat Margherita Pero
Karl Anthony Hribernik Mirco Peron
Hans-Henrik Hvolby Fredrik Persson
Natalia Iakymenko Marta Pinzone
Niloofar Jafari Fabiana Pirola
Tanya Jahangirkhani Adalberto Polenghi
Tim Maximilian Jansen Daryl John Powell
Yongkuk Jeong Rossella Pozzi
Kerstin Johansen Vittaldas Prabhu
Björn Johansson Hiran Harshana Prathapage
Bjørn Jæger Moritz Quandt
Ravi Kalaiarasan Ricardo Rabelo
Dimitris Kiritsis Mina Rahmani
Takeshi Kurata Slavko Rakic
Juhoantti Viktor Köpman Mario Rapaccini
Nina Maria Köster R. M. Chandima Ratnayake
Danijela Lalić Eivind Reke
Beñat Landeta Daniel Resanovic
Nicolas Leberruyer Ciele Resende Veneroso
Ming Lim Irene Roda
Maria Linnartz David Romero
Flavien Lucas Anita Romsdal
Andrea Lucchese Christoph Roser
Egon Lüftenegger Nataliia Roskladka
Ugljesa Marjanovic Monica Rossi
Julia Christina Markert Martin Rudberg
Melissa Marques-McEwan Roberto Sala
Antonio Masi Jan Salzwedel
Gokan May Adrian Sánchez de Ocaña
Matthew R. McCormick Kszysztof Santarek
Khaled Medini Biswajit Sarkar
Jorn Mehnen Claudio Sassanelli
Joao Gilberto Mendes dos Reis Laura Scalvini
Hajime Mizuyama Maximilian Schacht
Eiji Morinaga Bennet Schulz
Sobhan Mostafayi Darmian Marco Semini
x Organization

Sourav Sengupta Mehmet Uzunosmanoglu

Fabio Sgarbossa Bruno Vallespir
Vésteinn Sigurjónsson Ivonaldo Vicente da Silva
Marcia Terra Silva Kenneth Vidskjold
Katrin Singer-Coudoux Vivek Vijayakumar
Ivan Kristianto Singgih Gregor von Cieminski
Lars Skjelstad Paul Kengfai Wan
Riitta Johanna Smeds Piotr Warmbier
Selver Softic Kasuni Vimasha Weerasinghe
Per Solibakke Shaun West
Vijay Srinivasan Stefan Alexander Wiesner
Kenn Steger-Jensen Joakim Wikner
Oliver Stoll Magnus Wiktorsson
Jan Ola Strandhagen Heiner Winkler
Jo Wessel Strandhagen Jong-Hun Woo
Nick B. Szirbik Thorsten Wuest
Endre Sølvsberg Lara Popov Zambiasi
Iris D. Tommelein Matteo Zanchi
Mario Tucci Yuxuan Zhou
Ebru Turanoglu Bekar Iveta Zolotová
Ioan Turcin Anne Zouggar
Arvind Upadhyay Mikael Öhma
Andrea Urbinati
 Contents – Part III

Towards Next-Generation Production and SCM in Yard and

Construction Industries

Towards a Concept for Digitalized Yard Logistics—Outlining the

Next-Generation Features . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Jo Wessel Strandhagen, Marco Semini, and Erlend Alfnes

Requirement Analysis and Concept Design of a Smart Mobile Factory for

Infrastructure Projects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Patrick Dallasega, Andrea Revolti, Felix Schulze, Lorenzo Benedetti,
and Doré de Morsier

Management and Emerging Technology in Maritime Logistics:

A Lewin Force Field Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
Atle M. Christiansen and Kenneth Vidskjold

Streamlining the Execution of Maritime Commissioning with a Digital

Assistance System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
Tim Maximilian Jansen, Oliver Karl, Ahmed Elzalabany,
and Hermann Lödding

Transforming Engineer-to-Order Projects, Supply Chains,

and Ecosystems

Challenges and Opportunities of Software-Based Production Planning

and Control for Engineer-to-Order Manufacturing . . . . . . . . . . . . . . . . . . . . . 67
Patrick Bründl, Micha Stoidner, Huong Giang Nguyen,
Andreas Baechler, and Jörg Franke

Towards the Digital Factory Twin in Engineer-to-Order Industries: A Focus

on Control Cabinet Manufacturing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
Micha Stoidner, Patrick Bründl, Huong Giang Nguyen,
Andreas Baechler, and Jörg Franke

Has the Pendulum Swinged Too Far from Just-in-Time to Just-in-Case

in the Aftermaths of Covid-19?. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
Jenny Bäckstrand and Andréas Malmstedt
xii Contents – Part III

Integrating Lean, Agile, Resilient and Green Supply Chain Management

in Engineer-to-Order Contexts: Insights from Expert Interviews . . . . . . . . . . . 112
Antonio Masi and Margherita Pero

Investigating On-Site Production in Construction Using Decoupling

Thinking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
Petter Haglund, Joakim Wikner, and Martin Rudberg

Clarifying the Interface Between Construction Supply Chain and Site -

A Key to Improved Delivery Efficiency. . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
Farah Naz and Anna Fredriksson

Capability Building Blocks for Digital Twin Development . . . . . . . . . . . . . . . 154

Sonika Gogineni, Cansu Tanrikulu, Jörg Brünnhäußer, Kai Lindow,
and Heiko Witte

Underlying Mechanisms for Planning Engineering Capacity and Load in an

Engineer-to-Order Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
Nils-Erik Ohlson

Exploring Challenges in a Low-Volume Product Industrialization

Process – A Railway Case Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
Vésteinn Sigurjónsson, Jessica Bruch, and Anna Granlund

Performance Management Collaboration Between Companies Involved in

the Industrialised Housebuilding Order Fulfilment Process . . . . . . . . . . . . . . . 199
Wolfgang Grenzfurtner and Martin Rudberg

Industry 4.0 Application in ETO Companies: An Empirical Comparison . . . . . 213

Felix Schulze and Patrick Dallasega

The Resilience of an ETO Archetype to Demand Shocks . . . . . . . . . . . . . . . . 229

Yuxuan Zhou, Jonathan Gosling, Mohamed Naim, and Xun Wang

Modelling Supply Chain and Production Systems

A Location-Routing Problem: Last-Mile Delivery with Drop-Off Facilities

for Return . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
Sungbae Jo and Ilkyeong Moon

Cost Evaluation of a (Q, r, K) Inventory Model with Two Demand Classes

of Lost Sales and Backorders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
Zengxu Guo and Haoxun Chen
 Contents – Part III xiii

Business Models for Electric Vehicle Fixed Charging Station Infrastructure

with Commercial and Non-commercial Uses. . . . . . . . . . . . . . . . . . . . . . . . . 271
Hakan Erdeş and Saadettin Erhan Kesen

Implementation of a Quality Cost Management Model: Case Study from the

Textile Industry Sector. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
Bruno Barros, Cristina Rodrigues, Sérgio Sousa, and Eusébio Nunes

Optimal Production Planning of Ice-Food Under Production, Backordering

and Renewal Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
Syrine Guinoubi, Yasmina Hani, Marwa Hasni,
and Abderrahmane Elmhamedi

Sustainable and Economic Success Factors for Urban Consolidation

Centres of Last-Mile Delivery in the Netherlands . . . . . . . . . . . . . . . . . . . . . 318
Philip d’Agnolo and Adriana Saraceni

Automating Loading and Unloading for Autonomous Transport:

Identifying Challenges and Requirements with a Systems Approach . . . . . . . . 332
Tarun Kumar Agrawal, Robin Hanson, Farook Abdullah Sultan,
Mats I. Johansson, Dan Andersson, Gunnar Stefansson,
Konstantina Katsela, and Michael Browne

Optimal Class-Based Storage System with Diagonal Movements. . . . . . . . . . . 346

Kasuni Weerasinghe, Fabio Sgarbossa, and Giulia Fede

Algorithms and Models for Automated Replenishment of Store Shelves –

Exploratory Research. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 360
Abhinav Majumder, Shiyu Sun, and Vittaldas Prabhu

A Simulation Optimization Approach to Inventory Optimization in Supply

Chain Networks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374
Farzaneh Mahmoudi, Alireza Eshghi, Mohadese Basirati,
and Erfan Hassannayebi

Design of Reconfigurable Cellular Manufacturing Systems

with Alternative Routing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
Mehmet Uzunosmanoglu, Veronique Limère, and Birger Raa

Investigating the Sustainable Development of Charging Stations for Plug-in

Electric Vehicles: A System Dynamics Approach . . . . . . . . . . . . . . . . . . . . . 400
Mohammad Pourmatin, Amir Fayaz-Heidari, Moein Moeini-Aghtaie,
Erfan Hassannayebi, and Mohadese Basirati
xiv Contents – Part III

Pricing Strategy of Apparel Supply Chain Considering Traceability

Awareness of Consumers Driven by Blockchain . . . . . . . . . . . . . . . . . . . . . . 417
Wenjie Wang, Jinxia Zheng, Yazhou Liu, and Lei Xie

Advances in Dynamic Scheduling Technologies for Smart Manufacturing

Scheduling Algorithm Using Path Relinking in Different Search Paths for

Production Process with Crane Interference . . . . . . . . . . . . . . . . . . . . . . . . . 433
Takashi Tanizaki, Shonosuke Fujiwara, Hideki Katagiri,
and Takayuki Kataoka

Buffer Sizing and Route Scheduling for Reliable Autonomous Vehicle

Operations in a Dynamic Environment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 449
Inkyung Sung and Peter Nielsen

Beyond the Lab: Exploring the Socio-Technical Implications of Machine

Learning in Biopharmaceutical Manufacturing . . . . . . . . . . . . . . . . . . . . . . . 462
Erik Flores-García, So Hyun Nam, Yongkuk Jeong, Magnus Wiktorsson,
and Jong Hun Woo

A Constraint Programming Model for a Reconfigurable Job Shop

Scheduling Problem with Machine Availability . . . . . . . . . . . . . . . . . . . . . . . 477
Zahra Mehdizadeh-Somarin, Reza Tavakkoli-Moghaddam,
Mohammad Rohaninejad, Zdenek Hanzalek, and Behdin Vahedi-Nouri

Prediction of Residual Dye Using Machine Learning Algorithms for an

Eco-Friendly Dyeing Process . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 491
Whan Lee, Hye Kyung Choi, Seyed Mohammad Mehdi Sajadieh,
Sang Do Noh, and Hyun Sik Son

Applying Multi-agent Reinforcement Learning and Graph Neural Networks

to Flexible Job Shop Scheduling Problem . . . . . . . . . . . . . . . . . . . . . . . . . . 506
Seung Heon Oh, Young In Cho, and Jong Hun Woo

Enhancing Operations Planning and Scheduling in Dynamic Production

Systems by Using CLIP. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520
Julia Markert, Matthias Kerzel, Michael Variola, Dominik Saubke,
Stephanie von Riegen, Lothar Hotz, and Pascal Krenz

Data-Driven Analysis and Assignment of Manual Assembly

Production Lines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535
Changha Lee, Jongpil Yun, Goo-Young Kim, Junwoo Lim,
Sang Do Noh, and Yongjin Kim
 Contents – Part III xv

NSGA-II for Solving a Multi-objective, Sustainable and Flexible Job Shop

Scheduling Problem. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 548
Candice Destouet, Houda Tlahig, Belgacem Bettayeb,
and Bélahcène Mazari

AI Vision Use Case for Digital Twin WIP Tracking in Heavy Industry . . . . . . 563
Jang Won Choi, SungJae Cho, Binglu Li, and Jong-Hoon Park

An Improved Method of Job Shop Scheduling Considering Reworking

and Reprocessing Based on Proactive Approach . . . . . . . . . . . . . . . . . . . . . . 576
Eiji Morinaga, Kenta Teramoto, and Hidefumi Wakamatsu

Optimized Task Planning of Transfer Robots Using

Reinforcement Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591
Ji Whan Park and Sang Do Noh

Adaptive Traffic Signal Control for a Mixed Autonomous and Traditional

Vehicles by Agent-Based Digital Twin Simulation . . . . . . . . . . . . . . . . . . . . 603
Harry Lim, Minseop Go, Taekyu Lim, and Duck Young Kim

Data Preparation for AI-Assisted Video Analysis in Manual Assembly

Task: A Step Towards Industry 5.0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 619
Yongkuk Jeong, Magnus Wiktorsson, Donggyun Park, Jesper Gans,
and Linda Svensson

Reactive Flexible Job Shop Problem with Stress Level Consideration . . . . . . . 632
Ehsan Yadegari, Damien Lamy, and Xavier Delorme

Smart Production Planning and Control

Does Regulating Work-In-Process Increase Throughput and Reduce Cycle

Times? An Assessment by Lab Scale System Models . . . . . . . . . . . . . . . . . . 649
Matthias Thürer, Shan Shan Li, Can Yang, Ting Qu,
and George Q. Huang

Systems Thinking Approach for Production Process Optimization Based on

KPI Interdependencies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 662
Heiner Winkler, Susanne Franke, Felix Franke, Iren Jabs,
Daniel Fischer, and Matthias Thürer

Modeling of a Matrix Production System for Simulation to Predict

Material Demand . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 676
Daniel Ranke and Thomas Bauernhansl
xvi Contents – Part III

Data-Driven Production Logistics: Future Scenario in Two Swedish

Companies Based on Discrete Event Simulation . . . . . . . . . . . . . . . . . . . . . . 691
Masoud Zafarzadeh, Magnus Wiktorsson, and Jannicke Baalsrud Hauge

Setup Time Prediction Using Machine Learning Algorithms: A Real-World

Case Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 707
Alberto Locatelli, Manuel Iori, Marco Lippi, and Marco Locatelli

Simple Analysis of Planning Quality in Production Logistics . . . . . . . . . . . . . 722

Tobias Hiller, Lena Osterkamp, Lea Vinke, Patrick Holtsch,
Alexander Mütze, and Peter Nyhuis

Planning and Control of Maritime Commissioning: Planning Concept . . . . . . . 735

Nina Köster, Christopher Mundt, and Hermann Lödding

Requirements Planning in the New Normal: Comparison Between Reorder

Point Method and DDMRP . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 750
Beatrice Marchi, Ivan Ferretti, and Simone Zanoni

Towards Smart Maintenance and Integrated Production Planning . . . . . . . . . . 765

Julia Pahl, Harald Rødseth, and Jan Ola Strandhagen

Smart Production Planning and Control; Concept for Improving Planning

Quality with Production Feedback Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . 779
Mina Rahmani, Øyvind A. M. Syversen, Anita Romsdal, Fabio
Sgarbossa, and Jan Ola Strandhagen

Spare Parts Demand Prediction by Using a Random Forest Approach . . . . . . . 793

Joakim Andersson and Evangelos Siminos

Artificial Intelligence of Things (AIoT) Strategies for a Smart

Sustainable-Resilient Supply Chain . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 805
Hamed Nozari, Reza Tavakkoli-Moghaddam, Mohammad Rohaninejad,
and Zdenek Hanzalek

PPC Layout and Order Net – Visualization for a Rapid PPC Analysis
and Design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 817
Hans-Hermann Wiendahl

Introduction of an Approach for the Identification of Interfaces Between the

Factory Planning Process and Quality Management for an Optimized
Planning Result. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 832
Tanya Jahangirkhani, Ninja vom Stein, Peter Nyhuis, and Manuel Löwer
 Contents – Part III xvii

Production Scheduling Using Production Feedback Data; An Illustrative

Case Study. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 844
Mina Rahmani, Anita Romsdal, Øyvind A. M. Syversen,
Fabio Sgarbossa, and Jan Ola Strandhagen

Author Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 859

Towards Next-Generation Production
and SCM in Yard and Construction
Industries
 Towards a Concept for Digitalized Yard
Logistics—Outlining the Next-Generation
Features

Jo Wessel Strandhagen1,2(B) , Marco Semini2 , and Erlend Alfnes2

1 SINTEF Digital, Trondheim, Norway
[email protected]
2 NTNU—Norwegian University of Science and Technology, Trondheim, Norway

Abstract. Yards are industrial sites for production and servicing of ships and
offshore maritime installations, such as oil and gas platforms and modules, off-
shore windmills, and fish farms—all essential products in the maritime industry.
Although many yards are performing highly complex and technically advanced
production, there is still a need to bring the internal logistics of yards to a corre-
sponding level of advancement. Industry 4.0 technologies may answer this need,
and this paper presents a concept for digitalized yard logistics. The concept is
developed through a concept development activity inspired by design science
research, based on a multiple case study of 8 Norwegian yards, including ship-
yards and offshore construction yards. By mapping these yards, in particular their
current level of digitalization, we propose a concept grounded in practice. The
concept is built upon four main features of digitalized yard logistics: i) seamless,
digitalized information flow, ii) identification and interconnectivity of objects, iii)
digitalized operator support, and iv) automated and autonomous material flow.
The paper describes and visualize how currently available digital technologies
can be applied in the yard logistics context, to achieve those four main features.
The concept may be used as inspiration for moving towards the next generation
of yard logistics. The paper also addresses qualitatively the potential effects of
digitalized yard logistics on yard logistics performance. In this way, the paper
may serve as a starting point for more advanced and specific developments, as
well as possible realizations, of digitalized yard logistics systems.

Keywords: Yards · engineer-to-order · logistics · digitalization · Industry 4.0

1 Introduction
Yard operations can be classified as engineer-to-order (ETO) manufacturing operations.
With the ETO manufacturing approach, some design and engineering as well as purchas-
ing and physical production are performed after a customer order has been contracted
(Gosling & Naim, 2009). ETO manufacturing is sometimes called one-of-a-kind man-
ufacturing, as products that are designed and engineered based on a specific customer
order are often the only ones of their kind. The implications for the manufacturer, or the
© IFIP International Federation for Information Processing 2023
Published by Springer Nature Switzerland AG 2023
E. Alfnes et al. (Eds.): APMS 2023, IFIP AICT 691, pp. 3–18, 2023.
https://doi.org/10.1007/978-3-031-43670-3_1
4 J. W. Strandhagen et al.

yard, is that, since every product is designed and engineered based on the customer’s
requirements, it will never make a product in exactly the same way again. The yard, in
this case, is also a logistics hub receiving materials from a large number of suppliers, with
the challenging task of efficiently coordinating and managing these materials to have
them processed and installed on the end product. This has major implications for inter-
nal logistics and creates a dynamic, uncertain, and complex manufacturing environment
(Bertrand & Muntslag, 1993). These characteristics distinguish this type of manufac-
turing environment from more repetitive manufacturing environments. The need for
coordination of material and information flows is critical (Mello et al., 2017), and tai-
lored approaches are required for effective and efficient management of manufacturing
operations (Adrodegari et al., 2015). However, there is a lack of logistics solutions that
fit the ETO context (Zennaro et al., 2019).
Research on Industry 4.0 related to the ETO context has received growing atten-
tion, although it is still in an early phase (Cannas & Gosling, 2021; Zennaro et al.,
2019). Moreover, research on the application of Industry 4.0 to manufacturing logis-
tics indicates that new digital technologies are easier to apply in companies in which
the repetitiveness is high (Strandhagen et al., 2017). For less repetitive environments,
such as yards and other types of ETO environments, application of digital technolo-
gies seems more difficult. The high complexity, uncertainty, and dynamism created by
the characteristics of the ETO environment are believed to be key factors affecting the
applicability of digital technologies. On the other hand, the potential for improvement
if digital technologies are successfully adapted and applied should be correspondingly
large, as complexity and dynamism are exactly what digitalization is expected to manage
more efficiently. Accordingly, digitalization is expected to be a promising approach and
enabler of improved yard logistics performance. However, it is still not clear how digital
technologies can and should be applied in yard logistics.
There is currently an emerging research stream on Industry 4.0 in ETO (Cannas &
Gosling, 2021), and research on the application of Industry 4.0 technologies in the
specific context of ETO manufacturing is seen as a central part of future research in
the field (Zennaro et al., 2019). Nevertheless, existing research has considered only a
limited number of specific, technological applications for specific areas or processes
in yard operations. The digitalization of yard logistics is still at a superficial level, and
more empirically based research is required to identify the most relevant application
areas. Therefore, this paper will investigate how digital technologies can be adapted and
applied to move towards the next generation of yard logistics.

2 Related Work

Willner et al. (2016) conceptualizes four archetypes of ETO products: complex, basic,
repeatable, and non-competitive ETO. These are determined by two dimensions: annual
units sold (average number of units sold over a period of n years) and engineering
complexity (engineering hours per the average of annual units sold). Complex ETO
products are produced in lower volumes and with a higher engineering complexity, for
example, ships, oil platforms, and nuclear plants (Willner et al., 2016). Yard operations
fall within this ETO category, and is also characterized by large-sized, complex products
 Towards a Concept for Digitalized Yard Logistics 5

with deep product structures, manufacturing carried out as large projects in fixed position
layouts, a high level of customization, and highly integrated and overlapping processes.
Research have highlighted a lack of advancements of digitalization in ETO in general
(Zennaro et al., 2019), and this lack seems to apply also to yard logistics specifically.
There are a few articles addressing various aspects of digitalization of shipbuilding
in general. They are predominantly exploring the broad outlines of digitalization of
the shipbuilding industry (Beifert et al., 2018; Blanco-Novoa et al., 2018; Fernández-
Caramés et al., 2018; Jha, 2016; Joe & Chang, 2017; Munín-Doce et al., 2020; Para-
González & Mascaraque-Ramírez, 2020; Ramirez-Peña et al., 2019, 2020; Sanchez-
Gonzalez et al., 2019; Stanić et al., 2018), with only a few investigating the application
of digital technologies at yards. Nevertheless, there have been efforts to conceptualize
digitalization in yard operations, and Strandhagen et al. (2019) outlined a set of four
required features of a digitalized yard logistics system, which this paper aims to build
on:
• Seamless, digitalized information flow
• Identification and interconnectivity of objects
• Digitalized operator support
• Automated and autonomous material flow
In Strandhagen et al. (2019), these four features were suggested to address four
corresponding logistics challenges that characterize the yard logistics system: IT system
integration and sharing of up-to-date information; localization of materials, equipment
and tools; complex and information demanding work for operators; and manual material
handling and irregular and disrupted flows.

3 Research Approach

The research in this paper followed a qualitative approach, which is particularly use-
ful when seeking to understand real-world situations and their patterns and structural
features (Flick et al., 2004). Accordingly, the research targeted qualitative data, which
can be powerful for both discovering and exploring new ideas (Miles et al., 2014). In
order to understand the context, needs and requirements of yard logistics—as well as
to map the current state of digitalization—a multiple case study was used as the main
element of our research approach, based on the steps and principles for case research
(Yin, 2018). Due to Covid-19 travel restrictions, only Norwegian yards were considered
for the study. Fourteen Norwegian yards were identified as potential cases fulfilling the
inclusion requirements—manufacturing sites with operations that could be classified as
complex ETO and yard operations—and ultimately, eight yards were visited and included
in the study. Data were collected from interviews, direct observations, and existing doc-
umentation, and a case description for each yard was developed and organized. Through
this process, key insights from each case yard were generated, and commonalities and
differences between the yards emerged, allowing the unique patterns of each case to
be observed. Table 1 gives an overview of the cases involved in the study, with a short
description of the products built at the yard and the yard size in terms of its total area
and typical number of operators.
6 J. W. Strandhagen et al.

Table 1. Overview of the cases involved in the study.

Case Primary activity Yard size (m2 ) # of yard operators

A Oil platforms and modules, offshore wind 234 000 1200
platforms
B Floating offshore platforms, platform topsides, 250 000 800
onshore facilities for oil and gas processing,
offshore wind platforms
C Steel jackets for offshore platforms, offshore 650 000 482
wind jackets, subsea structures
D Smaller service operations on a range of 20 000 50
different types of ships
E Refurbishments, rebuilds, repairs, upgrades and 91 000 500
smaller service operations on different types of
ships, and some outfitting operation
F Outfitting and service operations on ships for 30 000 142
offshore, fishery and other types of specialised
vessels
G Outfitting operations on advanced and 75 000 300
customised ships for certain markets
H Outfitting operations on advanced and 20 500 90
customised ships for certain markets

In the extension of the case research, a concept development activity was carried
out. This concept development activity is inspired by the design science research (DSR)
method, which aims at developing generic knowledge from real field problems, with
generic designs as the core research product (van Aken et al., 2016). The research
product of DSR can eventually take the form of a construct, a model, a method, or
an instantiation (Hevner et al., 2004). In this paper, the concept development activity
utilized the previously described case studies, which explicate the challenges of yard
logistics and identify the requirements for digitalized yard logistics. Accordingly, the
case studies were the contextual or environmental foundation for the concept devel-
opment, providing an understanding of the particular field problems of yard logistics.
Similar as a design science activity, the concept development activity was connected
to the scientific knowledge base through being built on reviews of applicable digital
technologies. These informed and guided the concept development. Accordingly, with
an understanding of the yard logistics context through case studies, and a connection
to the state of the art of digitalization, a concept for digitalized yard logistics could be
developed.
 Towards a Concept for Digitalized Yard Logistics 7

4 Current State of Digitalization of Yard Logistics

To outline future steps in yard logistics digitalization and identify feasible solutions,
an overview of the current state of digitalization in yard logistics is necessary. This
was explored through a multiple case study, structured by four key aspects: technology
implementation, strategic emphasis on digitalization, resources and initiatives, and uti-
lization of IT systems. These aspects were combined to evaluate the current level of
digitalization at each case yard, as shown in Table 2.
The mapping reveals that yard logistics still has a long way to go to achieve com-
plete digitalization. While some yards belong to digitalized companies, overall, the
implementation of digital solutions in yard logistics is limited. Only a few yards have
implemented relevant technologies, including digital devices and small-scale AR/VR
solutions, which show promise. Other yards are conducting tests and investigations of
potential technologies.

Table 2. Current state of digitalization in yard logistics at the case yards.

Case Technology Digitalization Digitalization IT system use Digitalization

yard implementation strategy resources and integration level*
and initiatives of systems
A Wi-Fi Digitalization Dedicated Several IT Medium
throughout the is part of the resources and systems
yard, tablets for company internal initiatives supporting yard
work strategy. Pilot working on logistics, but
management, projects and digitalization only partly
AR has been proof of integrated.
tested in concept Information
assembly, pilot studies made available
projects on for operators in
materials digital format.
tracking Some digital
connection to
the yard floor
B App for work Digitalization Digitalization is a Several IT Medium
management, is part of the focus area, systems
development of company conducted supporting yard
digital twin in strategy. Pilot mappings/studies logistics, but
progress, projects and on digitalization only partly
investigating proof of and how it can be integrated.
camera concept applied in the Information
recognition for studies company made available
materials for operators in
identification digital format.
Some digital
connection to
the yard floor
(continued)
8 J. W. Strandhagen et al.

Table 2. (continued)

Case Technology Digitalization Digitalization IT system use Digitalization

yard implementation strategy resources and integration level*
and initiatives of systems
C App for work Digitalization Digitalization is a Several IT Medium
management is part of the focus area, systems
company conducted supporting yard
strategy. Pilot mappings/studies logistics, but
projects and on digitalization only partly
proof of and how it can be integrated.
concept applied in the Information
studies company made available
for operators in
digital format.
Some digital
connection to
the yard floor
D No Digitalization No specific Only basic ERP Very low
implemented is not part of resources for functionality,
digital the company digitalization with limited to
technologies, strategy no yard logistics
informally support. Many
introduced to manual
AR/VR for the administrative
inspection of tasks, manual
ships registration and
writing of lists.
No digital
connection to
the yard floor
E Tablets for work Digitalization No specific Basic ERP Low
management on not part of the resources for functionality, IT
a few previous company digitalization system
projects strategy supporting
basic
procurement
tasks. Manual
administrative
tasks. No digital
connection to
the yard floor
(continued)
 Towards a Concept for Digitalized Yard Logistics 9

Table 2. (continued)

Case Technology Digitalization Digitalization IT system use Digitalization

yard implementation strategy resources and integration level*
and initiatives of systems
F No No strategic No specific IT systems for Very low
implemented emphasis on resources for specific
technologies digitalization, digitalization, and applications,
although they they are involved but not
recognize the in research integrated to
need for project support yard
digitalization applications on logistics.
the topic Digital
information
printed on paper
before
distribution to
operators. Many
administrative
tasks are
performed
manually, with
manual
registration and
writing of
lists—and
manual checks
of them. No
digital
connection to
the yard floor
G No Digitalization No designated Partly Medium
implemented is somewhat positions for integrated
technologies. part of the digitalization. structure of IT
Investigating company Some smaller systems, to
the potential to strategy digitalization some degree
use AR in initiatives and supporting yard
research participation in logistics.
projects research projects Digital
information
printed on paper
before
distribution to
operators. No
digital
connection to
the yard floor
(continued)
10 J. W. Strandhagen et al.

Table 2. (continued)

Case Technology Digitalization Digitalization IT system use Digitalization

yard implementation strategy resources and integration level*
and initiatives of systems
H No particular No strategic No specific IT systems for Low
technologies emphasis on resources for specific
implemented digitalization, digitalization applications,
although they but not
recognize the integrated to
need for support yard
digitalization logistics.
Digital
information
printed on paper
before
distribution to
operators.
Partly used item
tagging system
for
identification of
materials,
however not
automated and
requires manual
scanning of tags
attached to the
materials. No
digital
connection to
the yard floor
* Digitalization level has been assessed using the following ordinal scale: Very low, low, medium-,
medium, medium +, high, very high.

Regarding IT systems, some yards have comprehensive, self-developed logistics

systems with substantial data, but they lack full integration with other systems, resulting
in manual administrative tasks. Most yards operate without any IT system for yard
logistics, relying on analog formats and extracting information from ERP systems.
The mapping of digitalization strategies concerns whether the digitalization is an
explicit part of the strategy of the company operating the yard—and whether this is
extended to yard logistics. Only four out of eight yards have digitalization as part of their
company strategy, indicating a lack of focus on yard logistics digitalization. Moreover,
the lack of resources dedicated to working on digitalization initiatives seen in most of
the yards may impede advancements.
 Towards a Concept for Digitalized Yard Logistics 11

Among the selected cases, fabrication yards show the highest level of digitalization
in yard logistics. This could be attributed to factors such as the size of the operating com-
panies, as larger enterprises tend to have higher digitalization levels (Buer et al., 2020).
Another possible factor relates to the sectors the yards serve. Offshore construction typ-
ically has higher profit margins than shipbuilding and is, therefore, likely to have more
resources available for company development initiatives such as digitalization efforts.
The empirical data clearly indicates the existence of significant barriers in terms of
implementation costs and obtaining top management approval, as well as challenges in
assessing the potential benefits of digitalization initiatives. Many yards struggle with the
financial burden and justification of such investments. Interviewees also highlighted the
difficulty of finding solutions suitable for the demanding physical environment of yards,
characterized by large unprotected outdoor areas and metallic objects that pose chal-
lenges for certain digital technologies like localization systems. Resistance to change
emerges as another potential barrier. While not directly observed, statements from inter-
viewees suggest that operators may exhibit reluctance to embrace new technologies
in their daily work during future implementations. However, this reluctance could be
attributed to the immaturity or unsuitability of the technologies.

5 Towards a Concept for Digitalized Yard Logistics

The concept is based on technologies that are available today and aimed towards realistic
implementations of digital technologies in an industrial context, i.e., what it could look
like in the foreseeable future). Strandhagen et al. (2019) identified four required features
of a digitalized yard logistics system, and in the concept development process these are
transformed to the four elements of the concept for digitalized yard logistics. The four
elements are shown in and described in detail in the following paragraphs.

5.1 Seamless, Digitalized Information Flow

Efficient yard logistics relies on efficient distribution of the information that is required
to execute yard logistics activities and make decisions. Especially, the close interaction
between non-physical processes, such as engineering and project management, and pro-
duction requires integrated IT systems for the efficient control and execution of the yard
logistics activities. There is a need for a seamless, digitalized information flow, where
all subsystems are integrated. Information should flow from higher-level IT systems
to the production floor whenever needed, providing access to real-time information.
The general purpose of such seamless, digitalized information flow is to make the rele-
vant information available for the executing actors. Key aspects of seamless, digitalized
information flow in yard logistics include:
• The supervisors receive up-to-date, digitalized information from higher-level sys-
tems, such as ERP and project management systems, regarding the next work pack-
ages to complete, operator availability, material status, and resource availability (facil-
ities, production halls/areas, transportation resources) required for the distribution of
work packages.
12 J. W. Strandhagen et al.

• Assigning a work package (by a supervisor in a control system) could potentially

activate the required actions (the information is sent) to pick and bring the material to
the place of use (e.g., booking transportation and giving information to the transport
equipment that will perform the transportation) and activate the provision of infor-
mation (work package description) to the operators the work package was assigned
to.
• The transportation operators (or automated and autonomous transportation equip-
ment), upon being assigned to an internal transportation job, receive information
regarding the correct items to pick, where to pick them from, and where to deliver
them.
• The production operators, upon being assigned to a work package, receive the infor-
mation required to execute the job, such as drawings, work instructions, and about
which items (both material and equipment) are to be used and their locations in the
yard.
• Warehouse operators receive the required information upon receipt of incoming
materials to the yard and upon receiving internal material requisitions for internal
supplies.
• Progress reports from operations are automatically and instantly sent in a digital
format to the relevant parts of the yard organization.

5.2 Identification and Interconnectivity of Objects

It is challenging to gain an overview of all the materials, equipment, and tools needed to
perform yard logistics activities. IoT, with objects equipped with sensors and actuators to
enable storing and sharing of information, have the potential to mitigate these challenges
by providing identification and interconnectivity. Identifying and interconnecting objects
in a facility would enable a highly integrated way of managing operations. The general
purpose of identification and interconnectivity is to provide a complete overview of the
yard’s materials, equipment, and tools. We consider two possible approaches to real-time
location of objects:
• Physical object tagging of:

– Materials—transmitting information about their location, status, etc. This infor-

mation should then be available for the relevant logistics systems (e.g., for picking
the correct items from storage, finding them without having to search).
– Transportation resources—enabling the networking of all transportation resources,
potentially improving the process of selecting resources for different transportation
jobs (e.g., booking of an available and close resource for a transportation job).
– Other equipment used by operators.

• Identification of objects through vision/recognition technology

– Cameras mounted on transportation equipment, building structures, operators’

helmets, drones, or other suitable places, to scan objects in order to identify them,
update location, view status, etc.
 Towards a Concept for Digitalized Yard Logistics 13

– The information acquired is transmitted to relevant logistics systems or used

directly by the transportation equipment, operator, or drone for its current task
(e.g., to pick the object it is looking for).

Irrespective of the technical solution, the ability to identify and interconnect objects
in the yard will present great opportunities regarding the management of the objects.

6 Digitalized Operator Support

In yard logistics, it is critical for the operators to receive timely and correct information
about the tasks to be performed, such as, drawings and work instructions. Digitalized
yard logistics should therefore include digitalized operator support where operators have
access to digital tools, such as smartphones, tablets, or similar, to view work instruction,
drawings etc. In this way, digital technologies are utilized to provide enhanced support
for operators, ensuring rapid and easy access to information. Key aspects of digitalized
operator support include:
• Work package descriptions available electronically on handheld devices, such as
tablets or smartphones.
• AR or VR solutions to support various tasks, including:
• Warehouse operations such as picking, where AR-based information can provide
enhanced information on where to find the correct item in the warehouse.
• Outfitting jobs, where AR solutions can be used to visualize the operators’ tasks.
For instance, the specific item to be installed on a ship can be projected—through
AR-glasses—showing the operator where it is to be installed.
• Digitalized solutions for operators and supervisors to report progress. This should
make the important activity of progress reporting as convenient as possible.

6.1 Automated and Autonomous Material Flow

With the comprehensive material flow at yards, great potential lies in making material
flow more efficiently. In yard logistics, digital technologies can bring autonomy and
automation to the physical flow of materials. Components, parts, assemblies, tools,
equipment, and other objects could then be transported more efficiently and with less
human intervention. Key aspects of automated and autonomous material flow in yard
logistics include:
• Automatic guided vehicles (AGVs), autonomous mobile robots (AMRs), and collab-
orative robots (COBOTS) for material handling and performing logistics operations
in warehouses.
• Automated and possibly autonomous conveyors, cranes, and vehicles (self-elevating
transporters, multiwheelers, etc.) for transporting heavy or high-volume materials
around the yard
• AMRs (vehicles or drones) for transporting light, low-volume materials around the
yard
• Automatic storage systems, such as Pater Noster material handling systems, for
efficient storage of smaller sized materials.
14 J. W. Strandhagen et al.

6.2 Concept Features and Visualization

Together, these four elements form a holistic concept for digitalized yard logistics.
Figure 1 shows ten features the concept can bring to yard logistics:
1. Digital product information from design and engineering to supervisors and
operators.
2. Cloud-based information management for yard logistics information, including
product information from design and engineering, progress information from
production, inventory information from warehouse, work package information, etc.
3. Supervisors equipped with digital devices with information relevant for work
management.
4. Digital assignment of work packages to operators, along with work package
descriptions and product information made available for operators on digital devices.
5. Interconnection of transportation equipment, receiving information on new jobs,
such as, when and where to pick up which materials and where to deliver them.
6. Identification and location of objects through physical object tagging or vision
technology.
7. Items in warehouses identifiable through technology and connected to work
packages based on availability and needs.
8. Autonomous material handling in warehouses and other storage areas.
9. Operators performing outfitting operations equipped with AR devices that provide
support during outfitting.
10. Yard equipment interconnected and digitally assigned to jobs, with digital commu-
nication of status.

Fig. 1. Visualization of the ten features of the concept for digitalized yard logistics.

The concept has been developed on the basis of technologies that are available
today, albeit not currently commonplace at yards. Accordingly, there are several tech-
nology requirements that are necessary for the concept to be realized, which include the
following:
 Towards a Concept for Digitalized Yard Logistics 15

• AR devices. There are several types of AR devices available today that may suit a
yard logistics context. A physical device, in the form of a smartphone, tablet, headset
or glasses, equipped with the required hardware and software to run AR applications
is necessary.
• Identification technology system, either based on physical object tagging, for exam-
ple a RFID system, or based on vision/recognition technology. This requires both
hardware and software.
• Autonomous vehicles and automation technology for autonomous and automated
material handling.
• Networking technology to transmit information wirelessly between systems, objects,
etc.
• Software for logistics control, including the control logic.
For the successful realization of such a heavily technology-based concept, the human
aspect of yard operations must be considered and addressed. Certain parts of the concept
build on operators’ adoption of new technologies, such as wearables and other digital
devices, in their daily tasks. Accordingly, this may require changes in the way the people
involved in yard logistics work. For the described concept, adaptation is needed with
regard to the use of AR devices, such as glasses and headsets, interacting with digital
interfaces (e.g., smartphones and tablets), and becoming accustomed to autonomous
vehicles operating in the yard.
Another important issue for the realization of such a concept relates to the investment
requirements. The mapping of the current state of digitalization indicates that there are
potential barriers related to the investment costs. With the high uncertainty in the yards’
current situations, it is associated with great risk to make any investments if they cannot
be covered through current projects. Moreover, the novelty of a technology may make
it difficult to estimate the potential benefits.
Although the economic benefits may not be easily quantified, it is possible to qual-
itatively discuss the potential effects of digitalization on yard logistics performance.
Potential effects of digitalization on yard logistics performance include:
• Digital information flow enhances internal order processing involving production and
warehouse
• Reduced put away cycle times for incoming materials due to rapidly, digitally
available information on materials and their destined warehouse locations
• Improved information quality due to enhanced, digital information exchange
• More rapid location of the materials to be picked
• Less time spent locating materials with enhanced localization through networking of
objects
• Enhanced transport efficiency through better overview of items, facilities, and
equipment.
• Better overview for operators, reducing the unnecessary time spent walking and
searching for items
• Increased productivity of operators, e.g., though solutions to assist in material picking,
reducing picking time
• Increased productivity of operators, e.g., through digitally available up-to-date work
package descriptions
16 J. W. Strandhagen et al.

• More efficient warehouse operations with automated solutions

• More efficient internal transportation through automated transportation
• Reduced put away cycle time due to automated material handling solutions in
warehouse
• Reduced order picking cycle times due to automated warehouse solutions

7 Conclusions
The paper has proposed a concept for digitalized yard logistics. The development of
a concept for digitalized yard logistics is an effort to extend the general conceptual-
izations of digitalization (Dalenogare et al., 2018; Fatorachian & Kazemi, 2020; Frank
et al., 2019) to the yard logistics context. The existing literature includes some partly
related conceptual descriptions, where Ang et al. (2017) present a general framework for
digitalized ship design and engineering, production, and operation, Stanić et al. (2018)
describe “shipbuilding 4.0”—a general concept regarding the digitalization of ship-
building, including shipyards, shipowner, suppliers, and other actors in the shipbuilding
supply chain, and Woo & Oh (2018) describe “digital shipbuilding”—a computer-based
production management concept for modeling and simulating stages of the shipbuild-
ing process. Accordingly, the concept described in this paper stands out because it
addresses the digitalization of yard logistics—a narrower scope than existing shipbuild-
ing concepts, and a wider scope than concepts focusing on modeling and simulating
shipbuilding. This can make it more useful to practitioners in their effort to apply dig-
ital technologies in yard logistics. The concept may serve as a starting point for more
advanced and specific developments as well as possible realizations of digitalized yard
logistics systems, which should be the aim of further work. Thus, the paper contributes to
expanding the field of digitalization of manufacturing and logistics to the context of yard
logistics and yard industries. This enhances the general understanding and knowledge
of the potential impacts of digitalization and widens the solution space for solving yard
logistics challenges.

Acknowledgements. The research presented in this paper was conducted while Jo Wessel Strand-
hagen was a Ph.D. candidate at the Department of Mechanical and Industrial Engineering at NTNU,
which financed the research. The paper was written while Strandhagen was employed at SINTEF
Digital, which financed the finalization of the paper.

References
Adrodegari, F., Bacchetti, A., Pinto, R., Pirola, F., Zanardini, M.: Engineer-to-order (ETO) produc-
tion planning and control: an empirical framework for machinery-building companies. Prod.
Plann. Control 26(11), 910–932 (2015). https://doi.org/10.1080/09537287.2014.1001808
Ang, J.H., Goh, C., Saldivar, A.A.F., Li, Y.: Energy-efficient through-life smart design, manufac-
turing and operation of ships in an Industry 4.0 environment. Energies 10(5) (2017). https://
doi.org/10.3390/en10050610
Beifert, A., Gerlitz, L., Prause, G.: Industry 4.0 – for sustainable development of lean manufactur-
ing companies in the shipbuilding sector. In: Kabashkin, I., Yatskiv, I., Prentkovskis, O. (eds.)
RelStat. LNNS, vol. 36, pp. 563–573. Springer, Cham (2018). https://doi.org/10.1007/978-3-
319-74454-4_54
 Towards a Concept for Digitalized Yard Logistics 17

Bertrand, J.W.M., Muntslag, D.R.: Production control in engineer-to-order firms. Int. J. Prod.
Econ. 30, 3–22 (1993). https://doi.org/10.1016/0925-5273(93)90077-X
Blanco-Novoa, Ó., Fernández-Caramés, T.M., Fraga-Lamas, P., Vilar-Montesinos, M.A.: A prac-
tical evaluation of commercial industrial augmented reality systems in an Industry 4.0 shipyard.
IEEE Access 6, 8201–8218 (2018). https://doi.org/10.1109/ACCESS.2018.2802699
Buer, S.-V., Strandhagen, J.W., Semini, M., Strandhagen, J.O.: The digitalization of manufacturing:
investigating the impact of production environment and company size. J. Manuf. Technol.
Manag. 32(3), 621–645 (2020). https://doi.org/10.1108/jmtm-05-2019-0174
Cannas, V.G., Gosling, J.: A decade of engineering-to-order (2010–2020): progress and emerging
themes. Int. J. Prod. Econ. 241, 108274 (2021). https://doi.org/10.1016/j.ijpe.2021.108274
Dalenogare, L.S., Benitez, G.B., Ayala, N.F., Frank, A.G.: The expected contribution of Industry
4.0 technologies for industrial performance. Int. J. Prod. Econ. 204, 383–394 (2018). https://
doi.org/10.1016/j.ijpe.2018.08.019
Fatorachian, H., Kazemi, H.: Impact of Industry 4.0 on supply chain performance. Prod. Plann.
Control 32(1), 63–81 (2020). https://doi.org/10.1080/09537287.2020.1712487
Fernández-Caramés, T.M., Fraga-Lamas, P., Suárez-Albela, M., Vilar-Montesinos, M.: A fog
computing and cloudlet based augmented reality system for the Industry 4.0 shipyard. Sensors
18(6) (2018). https://doi.org/10.3390/s18061798
Flick, U., von Kardorff, E., Steinke, I.: A Companion to Qualitative Research. Sage Publications,
Upper Saddle River (2004)
Frank, A.G., Dalenogare, L.S., Ayala, N.F.: Industry 4.0 technologies: implementation patterns
in manufacturing companies. Int. J. Prod. Econ. 210, 15–26 (2019). https://doi.org/10.1016/j.
ijpe.2019.01.004
Gosling, J., Naim, M.M.: Engineer-to-order supply chain management: a literature review and
research agenda. Int. J. Prod. Econ. 122(2), 741–754 (2009). https://doi.org/10.1016/j.ijpe.
2009.07.002
Hevner, A.R., March, S.T., Park, J., Ram, S.: Design science in information systems research.
MIS Q. 28(1), 75–105 (2004). https://doi.org/10.2307/25148625
Jha, S.K.: Emerging technologies: impact on shipbuilding. Marit. Aff. 12(2), 78–88 (2016). https://
doi.org/10.1080/09733159.2016.1239359
Joe, T., Chang, H.: A study on user-oriented and intelligent service design in sustainable computing:
a case of shipbuilding industry safety. Sustainability 9(4) (2017). https://doi.org/10.3390/su9
040544
Mello, M.H., Gosling, J., Naim, M.M., Strandhagen, J.O., Brett, P.O.: Improving coordination in
an engineer-to-order supply chain using a soft systems approach. Prod. Plann. Control 28(2),
89–107 (2017). https://doi.org/10.1080/09537287.2016.1233471
Miles, M.B., Huberman, A.M., Saldaña, J.: Qualitative Data Analysis: A Methods Sourcebook,
3rd edn. Sage, Upper Saddle River (2014)
Munín-Doce, A., Díaz-Casás, V., Trueba, P., Ferreno-González, S., Vilar-Montesinos, M.: Indus-
trial Internet of Things in the production environment of a Shipyard 4.0. Int. J. Adv. Manuf.
Technol. 108, 47–59 (2020). https://doi.org/10.1007/s00170-020-05229-6
Para-González, L., Mascaraque-Ramírez, C.: The six dimensions of CSR as a driver of key results
in the shipbuilding industry. Corp. Soc. Responsib. Environ. Manag. 27(2), 576–584 (2020).
https://doi.org/10.1002/csr.1821
Ramirez-Peña, M., Fraga, F.J.A., Sánchez Sotano, A.J., Batista, M.: Shipbuilding 4.0 index
approaching supply chain. Materials 12(24) (2019). https://doi.org/10.3390/MA12244129
Ramirez-Peña, M., Sánchez Sotano, A.J., Pérez-Fernandez, V., Abad, F.J., Batista, M.: Achieving
a sustainable shipbuilding supply chain under I4.0 perspective. J. Clean. Prod. 244 (2020).
https://doi.org/10.1016/j.jclepro.2019.118789
Sanchez-Gonzalez, P.-L., Díaz-Gutiérrez, D., Leo, T.J., Núñez-Rivas, L.R.: Toward digitalization
of maritime transport? Sensors 19(4) (2019). https://doi.org/10.3390/s19040926
18 J. W. Strandhagen et al.

Stanić, V., Hadjina, M., Fafandjel, N., Matulja, T.: Toward shipbuilding 4.0—an Industry 4.0
changing the face of the shipbuilding industry. Brodogradnja 69(3), 111–128 (2018). https://
doi.org/10.21278/brod69307
Strandhagen, J.W., Alfnes, E., Strandhagen, J.O., Vallandingham, L.R.: The fit of Industry 4.0
applications in manufacturing logistics: a multiple case study. Adv. Manuf. 5(4), 344–358
(2017). https://doi.org/10.1007/s40436-017-0200-y
Strandhagen, J.W., Buer, S.-V., Semini, M., Alfnes, E.: Digitalized manufacturing logistics in
engineer-to-order operations. In: Ameri, F., Stecke, K.E., von Cieminski, G., Kiritsis, D. (eds.)
APMS 2019. IAICT, vol. 566, pp. 579–587. Springer, Cham (2019). https://doi.org/10.1007/
978-3-030-30000-5_71
van Aken, J., Chandrasekaran, A., Halman, J.: Conducting and publishing design science research:
inaugural essay of the design science department of the Journal of Operations Management. J.
Oper. Manag. 47–48, 1–8 (2016). https://doi.org/10.1016/j.jom.2016.06.004
Willner, O., Powell, D., Gerschberger, M., Schönsleben, P.: Exploring the archetypes of engineer-
to-order: an empirical analysis. Int. J. Oper. Prod. Manag. 36(3), 242–264 (2016). https://doi.
org/10.1108/IJOPM-07-2014-0339
Woo, J.H., Oh, D.: Development of simulation framework for shipbuilding. Int. J. Comput. Integr.
Manuf. 31(2), 210–227 (2018). https://doi.org/10.1080/0951192x.2017.1407452
Yin, R.K.: Case Study Research and Applications: Design and Methods, 6th edn. SAGE, Upper
Saddle River (2018)
Zennaro, I., Finco, S., Battini, D., Persona, A.: Big size highly customised product manufacturing
systems: a literature review and future research agenda. Int. J. Prod. Res. 57(15–16), 5362–5385
(2019). https://doi.org/10.1080/00207543.2019.1582819
 Requirement Analysis and Concept Design
of a Smart Mobile Factory for Infrastructure
Projects

Patrick Dallasega1(B) , Andrea Revolti1 , Felix Schulze1 , Lorenzo Benedetti2 ,

and Doré de Morsier2
1 Faculty of Engineering, Free University of Bozen-Bolzano,
Piazza Domenicani 3, 39100 Bolzano, Italy
[email protected]
2 EuroTube Foundation, Neugustrasse 66, 8600 Dübendorf, Switzerland

Abstract. Infrastructure construction projects like roads, railways and tunnels

can face cost overruns, time delays and consequently the resistance from the pub-
lic to their implementation. An important aspect to be considered is to provide a
continuous material supply to the site to avoid construction interruptions. Tradi-
tionally, the production is organized in a centralized way with large production
quantities and long lead times negatively impacting the environmental footprint of
the project. On the other hand, the manufacturing industry is changing to decen-
tralized configurations with the aim to increase the proximity to the customer.
The concept of introducing non-location bound mobile factories for producing
important components directly on the construction site could increase efficiency
and responsiveness while at the same time improving sustainability dimensions.
The paper presents the requirements and a concept for a Smart Mobile Fac-
tory suitable to supply the material to infrastructure construction projects in a
sustainable way. The requirements criteria were developed in focus group ses-
sions with industry experts and scholars. The concept was modeled and analyzed
with Building Information Modeling (BIM) and the Digital Factory Planning
software visTable®. It presents part of the results of the joint research project
“SMF4INFRA – Smart Mobile Factory for infrastructure projects” between the
ETH of Zurich and the Free University of Bozen-Bolzano. The Smart Mobile
Factory concept was developed by considering a case study for the construction
of a hyperloop transportation system in collaboration with the research institution
EuroTube Foundation.

Keywords: Distributed Manufacturing · Construction · Mobile Factory · Digital

Factory · BIM

1 Introduction
Infrastructure projects like roads, railways and tunnels are frequently affected by budget
overruns and time delays [1, 2]. In linear construction projects, like road, railway or tun-
neling, the construction site moves as the project progresses. A specific aspect that has
© IFIP International Federation for Information Processing 2023
Published by Springer Nature Switzerland AG 2023
E. Alfnes et al. (Eds.): APMS 2023, IFIP AICT 691, pp. 19–33, 2023.
https://doi.org/10.1007/978-3-031-43670-3_2
20 P. Dallasega et al.

to be considered avoiding construction interruptions is to provide a continuous material

supply to the site. Traditionally, the prefabrication part (e.g. tunnel rings) is organized
in a centralized way and conceived for large production quantities with long lead-times
where the production planning is often forecast-driven. Moreover, production systems
are often planned for a one-time implementation and for mass production. Specifically,
in centralized production configurations the shipping could have a high negative impact
on economic and ecological sustainability dimensions [3]. As a result, manufacturing
is changing to decentralized configurations being near to the customer and autonomous
production systems aiming at a self-regulation [4]. Here, two emerging research areas,
digitalization and industrialization, are fostering this transformation. Digitalization con-
siders the use of Building Information Modeling (BIM), Digital Factory Planning (DFP)
and Digital Twins (DT) to support the construction execution and operation in a more
sustainable way [5, 6]. Industrialized construction considers the use of advanced prefab-
rication as well as on-site factory approaches to better control as well as decrease cost
and time throughout the execution of an infrastructure construction project [7, 8].
Usually, in on-site factories, activities like the preparation of material, fabrication,
assembly and finishing are performed [9]. Specifically, on-site factories are useful for the
following fields of application [10]: i) Long distance between off-site manufacturing and
the construction site; ii) High unpredictability and uncertainty of material supplies; iii)
Difficult transport of large or heavy components; iv) Little space to store components on
site requiring a production and delivery on demand; v) Requirement of JIT deliveries for
specific parts avoiding construction delays; vi) Unstable construction progress caused
by e.g. bad weather conditions where the planning of JIT-deliveries becomes difficult.
The main contribution of the paper is to conceptualize and design a Smart Mobile
Factory (SMF) to supply in a sustainable way the needed material for linear infrastructure
construction projects. The paper presents part of the results of the joint research project
“SMF4INFRA – Smart Mobile Factory for infrastructure projects” between the ETH of
Zurich and the Free University of Bozen-Bolzano.
Here, a mobile factory is defined as a production facility that can be moved, which is
composed of modular and flexible production systems that allow the adjustment of pro-
duction quantity and type according to specific conditions on-site. The term smart means
in this context, a virtual representation of the physical factory that enables a bi-directional
information flow between the real production system and the digital representation
allowing an advanced planning and control.
According to the review of literature, the concept of a SMF has been little applied to
the construction of infrastructure projects even if these projects would be very suitable
for its application. Therefore, in the paper we present the requirements and a concept
for a SMF suitable to supply the material to infrastructure construction projects in a
sustainable way. The SMF concept was developed by considering a case study for the
construction of a hyperloop transportation system in direct collaboration with the Swiss
research institution Eurotube foundation.
 Requirement Analysis and Concept Design of a Smart Mobile Factory 21

2 Literature Review
The transformation in manufacturing from mass production to increasingly individual-
ized products in smaller quantities has created the need for compact, flexible and scalable
manufacturing facilities [11]. Further economic trends such as globalization paired with
simultaneous adjustments to local needs [12], reshoring and securing supply chains
as well as sustainability concerns lead to geographic self-reliant fabrication units that
apply modularized and mobile manufacturing systems allowing a swift adaption of vol-
ume, functionality and customization [11]. Technological advancements, such as DTs,
3D Printing, Autonomous Robots or Factory Digitalization leads to substantial changes
in production systems. Production can be managed in a decentralized way as well as
distributed networks can be coordinated via Information Communication Technologies
(ICT) [13].
This smaller and flexible manufacturing design with closer proximity to the end cus-
tomer is called Distributed Manufacturing (DM) [4]. The manufacturing systems evolved
from centralized approaches such as assembly lines, over Lean Production Systems, as
well as flexible and later reconfigurable manufacturing systems to more distributed sys-
tems such as web- and agent-based manufacturing systems and recently cloud-based
systems [14]. DM benefit from emerging technologies for improved manufacturing pro-
cesses even at small scale and from modern ICTs [4]. Thereby, DM aims to increase
production efficiency utilizing the compounding effects of supply chain and value-added
networks in manufacturing. Traditional production, with its centralized, high-volume,
and long lead-time mode of operation, is increasingly being replaced by DM, which
is driven by decentralized, autonomous, and end-user activities [4]. The necessity to
be able to adapt to changes in demand, modify their equipment, and even change their
location requires production systems to be smart and flexible as well as mobile [3].
According to Matt et al. [11] various types of DM design can be distinguished based
on their evolution. The types range from simple decentralized and standardized produc-
tion structures (type 1), over modular and scalable configurations (type 2) which can
also be reconfigured for product and quantity changes (type 3) as well as “made smart”
via intelligent and digitally networked production systems (type 4). According to the
authors, special forms of DM include service models for industrial contract manufactur-
ing (type 5), on-site temporary mobile factories such as on construction sites (type 6),
and model factories operated independently as franchises in a network (type 7), com-
plemented by ‘Cloud Production’ which uses 3D Printing to produce parts and products
in an adaptable and flexible way (type 8).
In construction, a mobile factory or ‘factory on-site’ usually is the temporary use
of fully functional mobile mini-manufacturing plants or mobile production cells at the
place of demand or consumption, thereby saving delivery time and reduce transport costs.
Other suitable scenarios for on-site factories are infrastructure projects with long dis-
tances and relatively high logistical efforts and costs. Practical examples include mobile
concrete batching plants for the production of ready-mix concrete in truck mixers directly
on site. The literature discusses on-site factories that combine modern manufacturing
technologies in small facilities often with a reconfigurable design as a possible solution
to these scenarios [9, 11, 15]. However, concrete application examples for infrastructure
projects are missing in the scientific literature.
22 P. Dallasega et al.

Besides the reduction in transports (between factory and site), on-site factories enable
efficient production and assembly tasks near the actual construction site, often in a
safer and cleaner working environment [9]. Alix et al. [3] emphasize that DM can help
reduce shipping costs, which often have a significant impact to the overall product costs.
However, the factor mobility is a somewhat under-researched sector of the production
realm [16]. Turner et al. [17] proposed a framework for sustainable modular construction
utilizing a DM approach that applies intelligent technologies such as IoT-enabled sensors
and RFID to better assemble and disassemble modular components and recycle materials
after their usage.
In the research by [18] a transportable container is used to house a prototype for a
reconfigurable small-footprint manufacturing system. The production system integrates
3D Printing, CNC milling, precision assembly and cleaning processes using modular
parts that can be exchanged swiftly and use standard supply and control interfaces. Sim-
ilarly, Alix et al. [3] present a framework for a reconfigurable and mobile manufacturing
system that can be moved to different locations without a loss of performance. Benama
et al. [19] present the concept of mobile manufacturing considering reconfigurability
and mobility for a solar energy field manufacturing system which is able to cope with
variability in demand volume and product mix on-site. Similarly, Gee and Brown [20]
designed a prototype of a portable micro-factory stored in a standard shipping container
suitable to produce and assembly timber frame components on a temporary construction
site, which helps to address issues of cyclical demand fluctuations and lees need for
initial investments in fixed-location facilities. As a result, less space is needed on site,
the unit can work autonomously without the need for existing facilities and requires less
capital investment.
Rosarius and Garcia de Soto [21] present a concept for an on-site factory that com-
bines prefabrication techniques and traditional work environments, supported by Lean
principles, which help to improve throughput and productivity, of panelized wood com-
ponents. The concept was further simulated and showed a reduction in lead times, wastes
in production, as well as minimization of resource usage thereby increasing sustainability
for industrialized construction. Stillström and Jackson [16] analyze mobile manufactur-
ing concepts and emphasize the need for these systems to be reconfigurable, modular,
reusable, and standardized.
Summing up, the previous mentioned studies indicate that such concepts must be
tested with further industrial cases in other sectors [19]. Further research is required
to understand and integrate reconfigurable and mobile manufacturing systems within
local construction networks [22]. Especially, large infrastructure projects such as rails,
roads, and tunnels have only marginally been considered by the literature and need
to be analyzed to identify the requirements and concepts for such flexible and mobile
manufacturing systems.
To further accelerate economic growth and make use of emerging technologies,
DM can be enhanced with Intelligent Manufacturing System (IMS) which can even
further enhanced by real-time optimization via IoT, DTs and cloud computing [23]. DT
can help to optimize manufacturing processes by providing up-to-date operational data
representation of physical objects thereby supporting decision-making in DM systems
 Requirement Analysis and Concept Design of a Smart Mobile Factory 23

[24]. Park et al. [25] show the use of DT and IoT for a connected micro smart factory
thereby improving monitoring, tracking and decision-support.

3 Research Method
Literature Review. To identify possible requirements and establish the current state of
the art for mobile and flexible factories, a literature review was carried out. A keyword
search was performed in the Scopus database. More in detail, a keyword search was
applied which was extended by forward and backward searches on the topics examined
to identify relevant publications. Table 1 shows a summary of the applied search strings.

Table 1. Search strings used in the literature review.

No. Search terms Font size and style

1 Mobile smart factory AND (TITLE-ABS-KEY (“Mobile factories”) OR
Infrastructure TITLE-ABS-KEY (“Mobile factory”) OR
TITLE-ABS-KEY (“On-site prefabrication”)
OR TITLE-ABS-KEY (“Mobile
Manufacturing”) OR TITLE-ABS-KEY
(“Mobile on-site factories”) OR
TITLE-ABS-KEY (“Mobile on-site factory”)
OR TITLE-ABS-KEY (“Reconfigurable and
mobile manufacturing system “) OR
TITLE-ABS-KEY (“Mobile Supply Chains”)
OR TITLE-ABS-KEY (“Flexible
manufacturing system”) OR TITLE-ABS-KEY
(“reconfigurable manufacturing system”) OR
TITLE-ABS-KEY (“Smart Mobile Factory”)
OR TITLE-ABS-KEY (“ modularized mobile
factory”) AND TITLE-ABS-KEY
(“Infrastructure”))
2 Distributed Manufacturing AND (TITLE-ABS-KEY (“Distributed
Construction manufacturing”) AND TITLE-ABS-KEY
(construction))
3 Distributed Manufacturing AND (TITLE-ABS-KEY (“Distributed
Infrastructure manufacturing”) AND TITLE-ABS-KEY
(infrastructure))

The search for the terms mobile, smart factory and infrastructure resulted in 128
documents, for DM and construction in 37 documents, and for DM and Infrastructure
in 53 documents. Several inclusion criteria, such as articles only published in English
language and studies published in peer-reviewed journals and conference proceedings,
were used. Any duplicates were removed. All articles were screened according to title,
abstract, and then full text. Further, forward and backward searches (snowballing) were
applied, resulting in a final sample of 31 articles for an in-depth analysis.
24 P. Dallasega et al.

Focus Groups. The aim was to identify the requirements for a mobile, smart, and
flexible factory. Therefore, a three-step methodology was applied.
Step 1: Identification of main requirements and sub-requirements. In this step, the
literature was used to identify requirements already mentioned in the scientific works.
Step2: Forming of a focus group to select suitable requirements. In this step, a
focus group was defined, consisting of industry experts and researchers. The identified
requirements from the literature were discussed and new requirements were added to the
list. Functional requirements and design parameters were derived by using the Axiomatic
Design methodology [26]. Here, specific attention was paid to consider the independence
axiom when decomposing functional requirements and design parameters. Although, this
was not the focus of the research presented in this article. Several categories and sub-
categories were formed. Pairwise comparison was used to rank the criteria in a group
decision making approach. Thereby, different experts gave their opinion collectively.
Specifically, our focus group consisted of eight participants from the academia and
industry. The years of experience in the formed focus group ranges from 2 to 18 years.
Step 3: Validation of identified requirements and sub-requirements. The identified
and summarized requirements and sub-requirements were submitted to the expert groups
for their opinion and validation. Eventually, requirements were adapted to be more
suitable for infrastructure projects. The results are shown in Table 2.
Concept Design and Analysis. To develop the concept of the different smart and
mobile factory configurations, Building Information Modeling (BIM) and the software
visTable® were used. In the case study, the software Graphisoft Archicad version 25
was used. In addition to the structural and architectural elements of the factory, specific
machinery was modelled from technical drawings provided by EuroTube Foundation.
By using IFC families, characteristics and attributes of the individual machines were also
created. This information and data will later be used for the creation of a Digital Twin.
The digital factory planning software visTable® was used to design, optimize, evaluate,
and visualize different factory layouts in a collaborative way. In visTable® transporta-
tion networks and material flow relationships can be designed and evaluated. Therefore,
in our research the software was used to simulate the logistics efforts of various SMF
layouts.

4 Concept of a Smart Mobile Factory

The results of the focus group sessions are summarized in Table 2. It shows the identified
requirements and sub-requirements for a concept of a SMF for infrastructure projects.
The sub-requirements are explained in detail as following: (1) Allow easy assembly and
disassembly of the mobile factory: This main requirement includes the sub-requirements
considering the installation and dismantling efforts needed for setting up a mobile factory
and its equipment on-site. (2) Allow for adjustment of the factory: In this criterion,
modularity and adaptability sub-requirements are considered for the design of the mobile
factory concept. (3) Reduce carbon footprint for the factory: These requirements include
sustainability aspects such as usage of reusable material, the reduction of transports or
the reduction of energy consumptions. (4) Minimize logistics and intralogistics of the
mobile factory: it considers the use of sustainable and appropriate logistical means.
 Requirement Analysis and Concept Design of a Smart Mobile Factory 25

When defining the concept of the SMF an initial life cycle assessment was done.
Initially, the SMF was designed using precast concrete panels. However, this approach
proved to be unsustainable due to the high cost associated with transporting these ele-
ments and their limited flexibility for repositioning. As an alternative, a second version
of the factory was conceived using wood, specifically cross-laminated timber (x-lam).
While this option demonstrated an optimal life cycle, the material costs exceeded expec-
tations. Considering the industrial production environment proposed in this article, the
raw material expenses for wood could not be justified. In both presented solutions, it
is crucial to consider assembly and disassembly times, which always require the use
of crane trucks for support. On the other hand, the pneumatic solution offers notable
advantages such as flexibility in design, utilization of recycled sheathing for the curtain
walling, and swift assembly and disassembly with minimal environmental impact, as it
only requires a compressor. Additionally, the weight and ease of transport between loca-
tions also contribute to deeming the pneumatic solution as the most suitable. Although the
detailed discussion of the life cycle assessment values for the proposed architectural solu-
tions was not the focus of this research, it is evident that the pneumatic approach offers
significant advantages in terms of sustainability, flexibility, and ease of transportation.
In Fig. 1 the chosen SMF concept is visualized. The building envelope consists of a
pneumatic tensile solution that is stabilized by means of reusable transport containers.
Pneumatic structures are a special type of architecture which is characterized by the
slightly higher air pressure of internal surroundings. The idea of using pneumatic struc-
tures for building structures is somewhat dated. In fact, the first experiments date back
to the beginning of the 18th century [27]. Only with the advent of new materials stability
and durability could be guaranteed. Most applications of pneumatic tensile solutions can
be found for exhibition/industrial or emergency scenarios [28].
The entire membrane structure is stabilized with the help of compressed air, which
is pumped continuously or at regular intervals, using compressors. These types of struc-
tures are very cost-effective. Budiyanto et al. [29] assert that the advantages of using
these technologies are that such kind of structures can be built, dismantled and moved
to other locations easily, safely, quickly. By using independent energy sources (e.g. pho-
tovoltaic solar energy systems), the pneumatic tent can be directly inflated and deflated
on the construction site. Here, the requirement of an easy deployment of the factory
building was considered. The containers allow the transportation of the pneumatic tent,
the compressors used to inflate it and other important equipment. Here, the requirement
of a relatively easy assembly as well as disassembly of the factory building on site had
been considered. To reduce the carbon footprint of the assembly and disassembly of
the factory, green energy production equipment is used. Specifically, over the transport
containers photovoltaic panels are placed that can be used to inflate the pneumatic tensile
solution as well as in part to operate the production equipment.
Furthermore, the SMF concept was subdivided into two modules: Module 1 Pre-
fabrication and module 2 Assembly. The prefabrication module consists of processes
like mixture preparation, mould preparation, casting and curing. The assembly module
consists of assembling concrete shells, post tensioning and sealing. The concrete shells
are assembled to one tube module within the SMF and afterwards installed into the
infrastructure project on-site.
26 P. Dallasega et al.

Table 2. Requirement analysis of SMF for infrastructure projects.

No. Main requirement Sub requirement

1 Allow easy assembly / disassembly of Low effort for preparation of the site (soil
factory leveling)
Easy assembly and disassembly of factory
building
Easy installation and disassembly of
machinery
2 Allow adjustment of the factory Allow adjustment of production type
Allow adjustment of production capacity
Allow adaptations to the soil requirements
of the construction site
3 Reduce carbon footprint for the factory Permit reusable material (i.e., wood,
pneumatic, concrete)
Use “green” energy (i.e., solar, electrical
vehicles)
Reduce energy consumption (i.e., increase
insulation)
Reduce CO2 emissions (transport)
Allow possibility to control temperature
(heating/cooling)
4 Minimize logistics and intralogistics efforts Use available logistic infrastructure (rail,
of the factory roads, etc.)
Use sustainable means of transport
Transportation-friendly design of the
factory concept

Fig. 1. Concept of SMF with pneumatic tensile solution.

 Requirement Analysis and Concept Design of a Smart Mobile Factory 27

The SMF concept was split into two parts to allow a parallelization of the SMF
operation at different construction sites. As an example, the prefabrication of the tube
rings is done in location A, where also the drying time of around 7 days has to be
considered. Afterwards the tube rings are assembled to one tube segment, which will
then be transported and assembled on-site. Once the prefabrication is completed, the
module 1 is disassembled and transported to a new construction location B. In this way,
during the drying time of the tube rings in location A, the tube rings for location B can be
prefabricated. Similarly, the assembly of the tube rings in location A can be parallelized
with the drying of the tube rings prefabricated in location B. As a result, the drying time
can be parallelized with value adding activities reducing so drastically the lead time.
The two modules of prefabrication and assembly are installed, operated, disassembled,
and transported in an asynchronous way until all tube modules had been erected on-site.

5 Case Study
The case study consists of a SMF to support the execution of an hyperloop infrastructure
project. EuroTube Foundation [30] is a Swiss research center tasked to accelerate the
development of hyperloop as a sustainable high-speed technology for intercity transport.
Hyperloop technology, i.e., high-speed vacuum transport, is an innovative means of
transport which would allow people and goods to travel at the speed of an airplane with
the ecological footprint of a train. In the future, hyperloop routes have the potential to
replace short-to mid-haul flights in Europe and around the world, significantly reducing
emissions while providing passengers with the highest level of comfort and speed.
The solution that EuroTube developed originates from the Swissmetro project started
in the 1980’s [31] and then reproposed by Elon Musk in a white paper [32], focusing on
environmentally friendly goods and passenger transportation. The hyperloop’s propul-
sion system is provided by a linear electric motor powered by renewable energy sources.
Reduced pressure in the tube and magnetic levitation reduce the amount of energy losses
due to friction, providing an eco-friendly, energy efficient, and no emissions transport
system. Among the proposed solutions away from the original design is the use of con-
crete rings to build the infrastructure. This change of material allows lower production
costs and easier maintenance and management during the life cycle of the infrastructure.
The shapes and spaces required for the SMF were studied with the EuroTube team,
and coded as follows: (1) the prefabricated modules (containers) to house meeting rooms,
changing rooms, offices, toilets; (2) the storage area for raw materials and materials
(cement, steel, sheaths); (3) the mixing and concreting area, where the moulds are placed
to preform the reinforced concrete rings; (4) the curing zone to allow the components to
dry; (5) the area for assembling and tensioning the elements to form the complete tubes
(20 m in length); (6) the sealing zone to coat the tube with air-tight sheaths to ensure
vacuum.
To allow an easy adaptation of the SMF to on-site requirements, different shapes
were conceptualized. The different shapes were designed by using the BIM software
Graphisoft Archicad. The layout design, the material flow analysis and its optimization
were done by using the Digital Factory Planning tool visTable®.
Figure 2 shows an extract of the material flow analysis with visTable® of the SMF
concept with the pneumatic tensile solution and the so called “T-version”. Here, the
28 P. Dallasega et al.

prefabrication module is attached to the assembly module. The advantage of this config-
uration is that the SMF consists of one single structure and so less construction effort is
needed on-site. The shortcomings are that this configuration requires high transportation
efforts caused mainly by the transportation of the concrete rings with a forklift. More-
over, in the mould preparation area, the mixture trucks have to enter the factory building
(see mould preparation in Fig. 2), creating so air pollution as well as noise and impact-
ing negatively the social sustainability dimension. Another disadvantage is that it is not
possible that module 1 (prefabrication) and module 2 (assembly) work independently
increasing so the lead time of tube production during the operation phase of the SMF.

Fig. 2. Concept SMF pneumatic tensile solution – Version “T”

Figure 3 depicts the SMF pneumatic tent solution with the “version L” configuration.
Here, the strengths are that module 1 and module 2 are divided allowing so to use in a
more flexible way the available space where to place the factory modules. Furthermore,
as anticipated before, module 1 and module 2 can be operated independently allowing
so to exploit better the drying time. Moreover, compared to the version “T” it allows
less total transportation effort (9,88 km per year compared to 9,99 km per year). The
shortcomings are that the separation of module 1 and module 2 are divided and thus they
may increase their assembly effort on-site (compared to the version “T”). Moreover, the
storage area is organized as a conventional block storage in cramped conditions. As such,
concrete rings that are stored first cannot be retrieved first hindering so to implement
the so-called First In – First Out (FIFO) concept. Figure 4 displays the SMF pneumatic
tent solution with the “version S” configuration. In this version the concrete ring storage
area was improved by implementing a so-called pass-through storage system.
This allows to reach a FIFO storage and retrieval for the module 2 (assembly, post
tensioning and sealing). Furthermore, the pass-through storage system allows to reach
less overall transportation effort. The shortcomings are that this version requires higher
investment costs for the pass-through storage as well as more effort in assembling the
SMF on-site. Moreover, it requires also more space to erect the pass-through storage
on-site.
 Requirement Analysis and Concept Design of a Smart Mobile Factory 29

Fig. 3. Concept SMF pneumatic tensile solution – Version “L”

The quantitative comparison of the three concept versions is summarized below in

Table 3. Version S has the shortest transportation effort for the forklift, while the distances
for the transportation effort of the gantry crane are relatively similar.

Table 3. Quantitative Results of material flow analysis.

Concept Transportation Effort Forklift Transportation Effort Gantry Crane

T 9.33 km 662.06 m
L 9.24 km (−1%) 634.35 m (−4%)
S 3.39 km (−63%) 678.92 m (+7%)

To select the most suitable configuration, a scoring model (also called utility anal-
ysis) was used. This method systematically supports decisions through evaluation and
selection of optimal alternatives and allows the capturing of both objective and subjec-
tive information. It was chosen because the method is easy to understand and to use and
it allows the addition of further contemporaries after other contemporaries have already
been evaluated. Table 4 summarizes the qualitative results of the three concept versions
corresponding to the main requirements and sub-requirements defined in Table 2. The
rating scale for the criteria from 1 (very negative) over 3 (neutral) until 5 (very positive)
was chosen. The requirements were also weighted in a value analysis that was carried out
by all authors together. Version S has proved to be the best variant quantitative as well
as qualitative in our research. The reasons for this are the significantly lower transport
costs than with the other two variants and the improved storage system. Version S will
also be implemented for the test track and serves as the basis for the DT development.
The final model “S” that was selected is visualized in Fig. 5. The pneumatic structure
is anchored to containers, which in turn are fixed to foundations. The dimensions of the
factory dome are 12 m in height and 22 m in width. The dimensions were designed to
guarantee appropriate space for the production activities inside, considering maximum
possible shapes and tensions for the pneumatic structure. The pneumatic structure alter-
nates with spaces covered with simple curtains to allow simple openings in the roof for
30 P. Dallasega et al.

Fig. 4. Concept SMF pneumatic tensile solution – Version “S”

Table 4. Utility analysis of different SMF configurations

Main criteria Weighing Rating of Concept

Version T Version L Version S
Allow easy assembly/disassembly of factory 0.3 4 3 2
Allow adjustment of the factory 0.15 2 4 3
Reduce carbon footprint for the factory 0.2 2 3 4
Minimize logistics and intralogistics efforts of 0.35 3 4 5
the factory
Total 1.0 2.95 3.5 3.6

the insertion of elements and machinery. These gaps between the structure also allow a
division of workspaces.
 Requirement Analysis and Concept Design of a Smart Mobile Factory 31

Fig. 5. 3D model of concept SMF pneumatic tensile solution – Version “S”

6 Conclusion
In this article, we introduced the concept of a Smart Mobile Factory to mitigate costs and
environmental impact in linear infrastructure construction projects. A detailed literature
review examines the concepts of Distributed Manufacturing and Mobile Factory. By
means of focus group sessions, specific requirements for the development of a Smart
Mobile Factory to sustainably supply infrastructure construction projects were defined.
Based on the requirement analysis the SMF concept was developed, which consists
mainly of a pneumatic tensile solution that is stabilized by means of reusable transport
containers and two factory modules (prefabrication and assembly) to allow a continuous
operation while moving the SMF to different construction site locations.
Furthermore, three specific SMF layout types were modeled in BIM and analyzed
as well as improved by using the digital factory planning software visTable®. The three
SMF layout configurations (version “T”, “L” and “S”) were compared in a quantitative
and qualitative way by using the proposed requirement analysis. Thus, the pneumatic
tensile SMF concept with the version “S” configuration was selected as the one with the
lowest transportation effort and most able to satisfy the requirements for a Smart Mobile
Factory to sustainably supply infrastructure projects.
The main research limitations are that the assembly, disassembly and transportation
of the two factory modules - prefabrication and final assembly - were not analyzed from
a sustainability point of view. In further efforts, sustainability dimensions will be con-
sidered when defining the SMF locations. Here, a methodology based on Multi-Criteria
Decision Making (MCDM) as well as Fuzzy Analytical Network Process (FANP) will
be developed to support the planner in deciding the suitable locations where to place the
SMF by considering sustainability dimensions. Moreover, in future research activities
also other resource flows besides the material flows (e.g. the transportation of machinery
and equipment) will be considered. Furthermore, the developed digital model will be
transformed to a Digital Twin allowing a bidirectional information flow during the oper-
ation of the factory. Finally, based on the proposed SMF concept a physical prototype
will be built.
32 P. Dallasega et al.

Acknowledgements. The research presented in this article was carried out within the research
project “Smart Mobile Factory for Infrastructure Projects (SMF4INFRA)”, which has received
funding from the Autonomous Province of Bolzano-Bozen as a Joint-Project South Tyrol –
Switzerland 2021.

References
1. Verweij, S., Gerrits, L.M.: Understanding and researching complexity with Qualitative
Comparative Analysis. Evaluation 19(1), 40–55 (2013)
2. Flyvbjerg, B., et al.: Five things you should know about cost overrun. Transp. Res. Part A
Policy Pract. 118, 174–190 (2018)
3. Alix, T., Benama, Y., Perry, N.: A framework for the design of a reconfigurable and mobile
manufacturing system. Procedia Manuf. 35, 304–309 (2019)
4. Srai, J.S., et al.: Distributed manufacturing: scope, challenges and opportunities. Int. J. Prod.
Res. 54(23), 6917–6935 (2016)
5. Costin, A., Adibfar, A., Hu, H., Chen, S.S.: Building Information Modeling (BIM) for trans-
portation infrastructure–literature review, applications, challenges, and recommendations.
Autom. Constr. 94, 257–281 (2018)
6. Sacks, R., Brilakis, I., Pikas, E., Xie, H.S., Girolami, M.: Construction with digital twin
information systems. Data-Centric Eng. 1(6) (2020)
7. Wang, M., Wang, C.C., Sepasgozar, S., Zlatanova, S.: A systematic review of digital tech-
nology adoption in off-site construction: current status and future direction towards Industry
4.0. Buildings 10(11), 204 (2020)
8. Arashpour, M., Bai, Y., Kamat, V., Hosseini, R., Martek, I.: Project production flows in off-site
prefabrication: BIM-enabled railway infrastructure. In: ISARC (2018)
9. Martínez, S., Jardón, A., Víctores, J.G., Balaguer, C.: Flexible field factory for construction
industry. Assem. Autom. 33(2), 175–183 (2013)
10. Rauch, E., Matt, D.T., Dallasega, P.: Mobile on-site factories—scalable and distributed man-
ufacturing systems for the construction industry. In: International Conference on Industrial
Engineering and Operations Management, pp. 1–10 (2015)
11. Matt, D.T., Rauch, E., Dallasega, P.: Trends towards distributed manufacturing systems and
modern forms for their design. Procedia CIRP (2015)
12. Schmid, S., Grosche, P.: Managing the international value chain in the automotive industry.
Bertelsmann Stift, 105–120 (2008)
13. R. E, D. P, and D. Matt: Sustainable production in emerging markets through Distributed
Manufacturing Systems. J. Clean. Prod. 135, 127–138 (2016)
14. Wu, D., Rosen, D.W., Wang, L., Schaefer, D.: Cloud-based design and manufacturing: a new
paradigm in digital manufacturing and design innovation. CAD Comput. Aided Des. 59, 1–14
(2015)
15. Rahaman. M.: One Platform-different cities: an automated plug and play modular building
construction & city planning system. Adv. Constr. Build. Technol. Soc. 14–24 (2014)
16. Stillström, C., Jackso, M.: The concept of mobile manufacturing. J. Manuf. Syst. 26(3–4),
188–193 (2007)
17. Turner, C., Oyekan, J., Stergioulas, L.K.: Distributed manufacturing: a new digital framework
for sustainable modular construction. Sustainability 13, 1–16 (2021)
18. Adamietz, R., Giesen, T., Mayer, P., Johnson, A., Bibb, R., Seifarth, C.: Reconfigurable
and transportable container-integrated production system. Robot. Comput. Integr. Manuf. 53,
1–20 (2018)
 Requirement Analysis and Concept Design of a Smart Mobile Factory 33

19. Benama, Y., Alix, T., Perry, N.: Reconfigurable manufacturing system design: the case of
mobile manufacturing system. Sci. Arts Métiers (SAM) (2014)
20. Gee, S., Brown, A.: A mobile system for the on-site assembly of timber frame components:
the development of an agile, low-cost alternative to offsite prefabrication. Sustainability 14(2)
(2022)
21. Rosarius, A., De Soto, B.: On-site factories to support lean principles and industrialized
construction. Organ. Technol. Manag. Constr. 13, 2353–2366 (2021)
22. Wagner, H.J., Alvarez, M., Kyjanek, O., Bhiri, Z., Buck, M., Menges, A.: Flexible and
transportable robotic timber construction platform. Autom. Constr. 120 (2020)
23. Nie, Q., Tang, D., Liu, C., Wang, L., Song, J.: A multi-agent and cloud-edge orchestration
framework of digital twin for distributed production control. Robot. Comput. Integr. Manuf.
82(10), 25–43 (2023)
24. Sahal, R., Alsamhi, S.H., Brown, K.N., O’shea, D., McCarthy, C., Guizani, M.: Blockchain-
empowered digital twins collaboration. Machines 9, 193 (2021)
25. Park, K.T., et al.: Design and implementation of a digital twin application for a connected
micro smart factory. Int. J. Comput. Integr. Manuf. 32, 596–614 (2019)
26. Rauch, E., Matt, D.T., Dallasega, P.: Application of axiomatic design in manufacturing system
design: a literature review. Procedia CIRP 53, 1–7 (2016)
27. Chi, J.Y., Pauletti, R.M.O.: An outline of the evolution of pneumatic structures. Paper
Presented in II Simp. Latinoameric ano Tensoestructuras (2005)
28. Iqbal, M., Budiyanto, H., Bonifacius, N.: The parameters of emergency disaster hospital
with inflatable pneumatic structure. In: 5th International Conference Graduate School on
Sustainability (2020)
29. Budiyanto, H., Winansih, E., Setiawan, A.B., Setiawan, M.I.: Portable stage and pneumatic
air inflated roof structure with independent energy as a means of exhibition of SME products.
Int. J. Sci. Eng. Res. 6(9), 48–51 (2018)
30. Eurotube (2023). https://eurotube.org/
31. Jufer, M., Perret, F.-L., Descoeudres, F., Trottet, Y.: Swissmetro, an efficient intercity subway
system. Struct. Eng. Int. 3(3), 184–189 (1993)
32. Musk, E.: Hyperloop Preliminary Design Study Technical Section (2013)
 Management and Emerging Technology
in Maritime Logistics: A Lewin Force Field
Analysis

Atle M. Christiansen(B) and Kenneth Vidskjold(B)

University of South-Eastern Norway, Post Office Box 4, 3199 Borre, Norway

{ac,kv}@usn.no

Abstract. When working with research and development [R&D] projects involv-
ing emerging technologies in transportation, many challenges regarding testing
and implementation of new products and processes will materialise. Not only the
qualification of technology and humans involved, but also conservative markets
and customers, national and international rules and regulatory bodies that admin-
ister them will resist change and seek to maintain status quo. To provide new
knowledge and improved project execution in R&D organisations we have con-
ducted a Lewin force field analysis utilising an extended version of Lewin’s initial
framework. The input data for this analysis are the result of four years of participa-
tion in a typical triple helix project on seaport development, focusing on autonomy
of vessels, automation in seaports, shore power and general electrification of equip-
ment and transportation. Despite all good intentions and technological prowess,
many of these projects fail to prove feasible and fall short of improving not only
efficiency but also safety and security. The aim of this paper is twofold: (i) Present
Lewin’s force field analysis as a tool for providing input to the project planning and
execution process. (ii) Identify drivers and barriers for successful projects based
on the knowledge and practice gained in R&D collaboration projects involving
maritime transport actors. As a result of this research seven major drivers and
barriers to project execution were identified and discussed.

Keywords: Automation · Autonomy · Lewin Force Field

1 Introduction
For obvious reasons we endeavour to conduct research and development projects that
end on schedule, on budget and successful. In the case of innovation projects this can
be a challenge. Innovation can be both a process and an outcome. When dealing with
projects involving both technological and societal changes we must relate to both aspects
of innovation. Our preliminary investigations revealed that projects dealing with both
technology and society are especially challenging and intricate. Technology may be
readily available; all rational and conventional tools of analysis show the feasibility and
that there is an obvious demand for these projects to succeed. Nevertheless, they some-
times fail. Planning and project execution may be flawless, but the desired deliverables
© IFIP International Federation for Information Processing 2023
Published by Springer Nature Switzerland AG 2023
E. Alfnes et al. (Eds.): APMS 2023, IFIP AICT 691, pp. 34–48, 2023.
https://doi.org/10.1007/978-3-031-43670-3_3
 Management and Emerging Technology in Maritime Logistics 35

fail to materialise. Many of these projects deal with green transformations and increasing
sustainability within transportation. Since there are many national and global policies,
regulations and legislation applicable this can halt project progress and hinder success.
The results of failure are fewer options and a delay in the vital shift towards a more
sustainable world. Project planning and execution are dynamic processes requiring con-
sistent supervision and assessment from stakeholders. Goals and intentions are mostly
clear and distinct, but often there is a lack of a coherent identification of the barriers and
drivers on the path towards achieving such targets. Our experience during these projects
and our preliminary investigations indicated that initial assumptions about some key
stakeholders were flawed, this provided the grounds for our main hypothesis:
Our main hypothesis is that project owners and stakeholders alike have faulty
assumptions about the goals and intentions of bodies upholding rules and regulations.
As a result they develop sub-optimal plans regarding qualification of technology and
new operations. A comprehensive force field analysis will elaborate the interaction of
different stakeholders in these projects. Traditional stakeholder analysis can often deal
more about the collection of knowledge about the goals, hierarchy and identification of
the stakeholders and less about their actions and interaction [1]. Other popular tools of
analysis, such as analytical hierarchy process, can have limited use in dealing with these
complex relationships [2, 3]. Also from our preliminary investigations, we detected a
need to identify some of the forces behind these faulty assumptions:
Our secondary hypothesis is that there is a need to find common theoretical perspec-
tives between research of technology and research of social topics. Positivists may fail
to realize that “truths” can be negotiated, change or falsified [4, 5]. Non-positivists may
disregard the limitations or feasibility of technology.
We aim to identify the restraining forces in the projects with the potential to be
manipulated and changed. According to Lewin’s theory of force fields, such forces can
sometimes be inverted to the benefit of the desired change in the projects.
This paper is organized with separate theory, method, results and discussion chapters.
In the two latter we are presenting the findings and discussing all five investigated projects
in common first because they were very similar, and then some more in-depth findings
and discussions are presented. At the end we make some general conclusions and present
some suggestions for further work.

2 Theory
2.1 Lewin’s Force Field Analysis
Traditionally, triangle models have been popular for visualising and understanding inter-
dependencies within projects. Most project managers know the iron triangle model:
Time, scope and money, and often use it to plan and keep the projects on track [6]. Crit-
icism can be raised about the usefulness of these models due to their over-simplification
of the relevant situation [7]. A more comprehensive, albeit more intricate and complex,
is Lewin’s force field analysis [8]. Its complexity has led to the use of an oversimplified
version of the model, especially within the consultancy and teambuilding business [9].
During his initial work in social psychology, Lewin realised that researchers needed to
not only explain the world but also provide solutions on how to take action and improve
36 A. M. Christiansen and K. Vidskjold

it. Hence the name action research which Lewin pioneered. Lewin was a pragmatist
and derived his methodology from C.S. Peirce; “… a dialectic process seeking best fit
or concordance and an interpretive (of many social perspectives) epistemology melded
into a quasi-experimental orientation” [10]. Lewin and his students developed tools to
facilitate this methodology and force field analysis is one of them. In our understanding,
field analysis has very much in common with case study especially as advocated by
Eisenhardt [11, 12], and proves very useful in case study research. Lewin conducted
many case studies utilising field analysis, inter alia, his publication on conflict solving
in industry [13]. Figure 1 is a part of an illustration from Lewin’s seminal paper on force
field analysis [14]. Lewin’s force field model can be difficult to grasp and understand
and his theory is essentially mathematically explained. There are miscellaneous inter-
pretations and simplifications to the model [9], and a full explanation of the model and
Lewin’s theories is far beyond the scope of this article. We perceive it as a 3D model with
a circular space disk (horizontal or x-y plane) in time (vertical or z-axis) thus illustrat-
ing, as Lewin names it, space-time. The disk represents the present field, and the future
is downwards. Hence the past is upwards. The whole horizontal plane is the life-space.
This is in time t and not dependent on a previous t-n. Figure 3 in the results section shows
a (x,y) force field with axes and forces assigned. If the absolute value of a force (vector)
in the field is F and the angle between vector and the axis x of desired change is ϕ, the
active and reactive forces in the field can then be calculated using simple trigonometry.
Following this; forces parallel to the x-axis (ϕ = 0°) will be purely active and forces
parallel to the y-axis (ϕ = 90°) will be purely reactive. Active forces will contribute in
their entirety to the desired change and reactive forces will have no contribution. Purely
active and reactive forces will probably exist only in theory.

Fig. 1. Force field model

Lewin is also adamant about not using the term - or put any meaning to “reaching
a goal” but rather “reaching a desired change” [15]. On this basis, Lewin defines a
“change” as a move of a point from x to x 1 resulting in a Δx. Depending on the direction
of the field surrounding (x,y) this will also result in a new y-coordinate y1 . Provided
this happened in space-time during the time interval Δt, the ratio Δx/Δt will determine
the change velocity. If there is no change it follows that the sum of all active forces is
zero and the change velocity is zero. Lewin attributes this to Quasi-Stationary Processes
 Management and Emerging Technology in Maritime Logistics 37

or Social “Habits” [16]. Such inner resistance needs additional forces to accomplish
desired change.

2.2 Seaports

Guido Weigend’s definition of a port from 1958 still holds: “The port is the place of
contact between land and maritime space, and it provides services to both hinterland and
maritime organisations” [17]. 80% of all transport in the world is by sea [18]. All sea
cargo is handled in a port at least twice. In a global perspective, the port holds many of the
keys to unlock a wider use of autonomy, automation and general digitalisation within
logistics. Among the miscellaneous operations and services carried out in the ports,
the essential services are the modal shifts of cargo, temporary or permanent storage and
general logistics. There is a great potential for emerging and new technology in seaports,
as this will improve the ratio of cost and energy use versus units handled [19, 20]. How
a port positions itself in the supply chain is commonly driven by how it is influenced
by public or private interests. The port can be fully controlled by the government as a
state port or it can be fully controlled by a private company as a private port. Typically,
ports will be positioned somewhere in both the public and private “influencing space”.
Historically, the ports have been under governmental control, but for several decades the
trend has been steadily growing towards more private control of their operations.
Port Management Models. Several factors influence how a port is organised, structured
and managed, Categorically, there is the public service port, the tool port, the landlord port
and the fully privatised port or private service port [21]. These models are distinguished
by certain characteristics, namely, to what extent the ports have their provisions from
public and/or private services; if the ports have a local, regional or global orientation;
ownership of the infrastructure; ownership of the superstructure and equipment and the
status of dock labour [21]. Seaports in Norway and Denmark are currently in the middle,
between state and private interest and with a mixed governmental and private control.
Most of the ports in our research projects were organised as a “landlord model”, which is
adopted for all ports in this study. The ports infrastructure and some of the superstructure
are controlled by the local municipality. The port operating company, facilitating port
labour, is normally privately controlled. A key element in the landlord model is the
balance of power in the decision making, i.e. the public and the private stakeholders are
sharing the responsibilities and commitments for the success of the port [21].

3 Method
At the University of South-Eastern Norway, a research group has conducted triple helix
type research and development projects within maritime transport for about a decade.
This study has selected four projects from maritime transport, namely, two large port
development projects, one autonomy project connecting an inland port to a seaport and
one loosely connected cluster of projects about urban waterways with a weak coupling to
ports. The rationale for this sampling was to provide for diversity and a heterogenous data
set. Of a total N = 15 ports, n = 3 was selected for further study. This sampling provided
38 A. M. Christiansen and K. Vidskjold

the optimal diversity and useful input to the Levin force field analysis. Our method for
data collection is a mixed method document analysis with additional qualitative data
gathering to follow the information needed to prove or disprove our hypothesis [22,
23]. In some cases, we gathered personal experience from either researchers or other
project participants conducting similar projects in typical triple helix constellations. In
these cases, we have made clear references to this in the text. Our initial data collection
was based on a set of assumptions due to the need to identify the strong or intermediate
forces, since the weaker ones were of less interest (Table 1).

Table 1. Data Sources

Project Goals Methods

Port research More sustainable port operations Selected 3 Scandinavian seaports
Observation, Interviews, Document
analysis
City Ferry Sustainable public transport Selected 2 Norwegian municipalities
Observation, Interviews, Document
analysis
Ship X Develop a fully autonomous ship One Scandinavian ship
Interviews and Document analysis

The research design followed Lewin’s outline and structures in his topological psy-
chology and field analysis in particular [5], and provided a comprehensible theoretical
perspective considering this is research having both human factors and technology per-
spectives initially [24]. Kurt Lewin was a pragmatist rather than a positivist. Clem Adel-
man sees Lewin’s clear influences by the American philosopher C.S. Peirce [10]. Peirce
is also the pioneer of abductive reasoning. In our view, Lewin was finding theory and
answers to questions through abduction rather than induction or deduction. Abduction
is about finding a warrant to explain an observed change to a given situation. Logically
it is formulated as; If X => Y and Z => Y then X => Z. If more different instances
warrant Y, further reasoning may apply and Occam’s Razor can be helpful. We con-
ducted an incremental abductive analysis to identify the different forces influencing the
projects we studied, and initially these would be designated as driving and restraining
forces (barriers) to the process of achieving the desired change. We were careful not to
apply super-position as the studied force field could not be proved to be a linear system.
Figure 2 makes a simple illustration of our research design. All projects and deliverables
in the studied projects are made anonymous.

Assumpons Force field Theorizing

Data More data
and analysis More realisc
gathering gathering
hypotheses (Iterave) world-view

Fig. 2. Lewin force field analysis as a scientific method

 Management and Emerging Technology in Maritime Logistics 39

4 Results

The results will be divided into three parts. Firstly, we will present the identified drivers
and barriers to desired change that we identified in the projects studied using the Lewin
FFA, and these are compared with the initial assumptions in the project plans and project
execution phase. Secondly, we will present the identified theoretical perspectives among
different project members and stakeholders. Thirdly, we will elaborate on the results
from the different domains.

4.1 Drivers and Barrier for Desired Change Common to All Projects

Many forces were identified, both driving and restraining, through our analysis. Very
small and minute forces probably occur in nearly infinitesimal numbers, and it is of
little use trying to find and identify them. No purely active or reactive forces were
discovered; all were composite. The closest we could identify as a purely active force
was policy. Public policies, strategies, international agreements and international- or
national conventions can be very strong drivers for change, but weak or even useless if
not enforced or honoured. Donald Trump’s withdrawal from the Paris Climate Agreement
is a memorable example. Another example of policy as a driver we could identify was
that all seaports, municipalities and regions had the UN Sustainable Development Goals
central to their strategies. In Table 2 we have compiled a list of the seven major forces that
we could identify in our data. At this point we will avoid deconstructing these, instead,
we will scrutinise this in the discussion chapter. We have, however, assigned them the
properties internal or external to the projects. In Fig. 3 we have made the corresponding
FFA diagram.

Table 2. 7 major field forces in the projects (Forces can be both drivers and barriers)

Identifier Field force Project

a Trend External
b Predictability and risk aversion (Governance) Internal
c Public opinion External
d Sustainable mobility External/Internal
e Market External
f Technology Internal
g Policy Internal/External

Both authors have experience working with R&D and building projects as an industry
partner or in industry alone. This enabled us to compare our findings within the industry
part of the helix, and not only from the outside in R&D projects like these [25], since
we both conducted academic studies into the projects investigated. When we compared
our experiences with these external projects to our current findings in R&D project work
40 A. M. Christiansen and K. Vidskjold

within academia there were significant differences. Building and construction projects
differ widely from the projects we research in this context. However, from a management
perspective this is interesting and there were many commonalities identified. In general,
industry partners are very flexible and adapt quite easily to changes and discrepancies
in projects and project plans. The 2019 Covid Pandemic made this especially relevant.
Loyalty to both organisation and task is strong among industry participants, and this
was identified as one of the strong drivers in the projects studied. Academia follows the
rules, governance makes and applies them. Their coupling to the public sector, or just
being part of it, makes them bureaucratic and contributors to restraining forces [26].
This was very visible in the application and funding process of these projects. The more
deterministic and more rigid the project execution, deliverables and goals we determined
in the project applications, the “better” it is considered by the funding body committees.
In the long run this leaves little room for change in the projects themselves. This results in
the project managers emphasising on a waterfall project execution the focus on goals and
not on what is actually happening in the projects - in stark contrast to what Kurt Lewin
is advocating. The purpose of this paper was not to assess FFA as a project management
tool but to use it to analyse the workings of past projects, hence we will leave these as
opportunities for future research.

Y:axis of reacve force

Z:pas me

a c e
f g
b d

-Z: future me X:axis of acve force

Fig. 3. Lewin force field example

4.2 Theoretical Perspectives

All projects studied were collaboration projects with governmental, academic and indus-
try partners. Triple helix and the later amendments to the model form a new constellation
in many settings, and we found nothing in these projects to indicate that the constellation
is not there, and that the collaboration does not work on an initial or parent level. As we
understand, the purpose of the triple (or more) helix model is to facilitate the emergence
of a shared understanding of the common life-space for all participants and stakeholders
[27–29]. At this point we encountered a problem because the different participants did
not share the same scientific perspectives. Even within one solitary axis, science itself,
we identified many divergent perspectives and standings within the same project.
In Table 3 we detail some of the different standings we identified, and this list is by
no means exhaustive. One of the great challenges in the social sciences today is to build
better models and answer more “what is it?” questions [30]. Models must be built on an
 Management and Emerging Technology in Maritime Logistics 41

epistemology and a theoretical perspective. It is challenging to build single models on

differentiating scientific perspectives [24].

Table 3. Examples of identified major scientific dependencies

Science Epistemology Theoretical perspective Axiology

Natural sci Objectivism Positivism None articulated
Computer sci Objectivism Positivism Technology will save us
Economic sci Subjectivism Postmodernism Monetary
Organisational theory Constructivism Pragmatism Sensemaking
Values become

Quite often we identified periods of no change in the projects studied. An FFA utilis-
ing abduction is a very useful tool for identifying the reasons behind these and facilitate
change. There is no progress in any project without some sort of change to the field and
eventually the life-space [31]. This lack of change was also observed several times in our
sampled projects, and as Lewin previously stated, these are a result of equilibriums that
occur in the projects and the project teams. Equilibriums can be the absence of driving
forces, such as lack of motivation or work initiative. We identified these quite often
among academic staff, and these again had connections with inflation due to overopti-
mism and or under-analysis in the project planning period. The task equivalent to a FFA
in “traditional” project planning is risk analysis [RA]. Risk probabilities and risk impacts
are identified and the ones with the highest probability and impact occurring together are
identified and avoided, mitigated or dealt with in some other way. Our experience, and
again here we are drawing on experience from previous projects, is that the analyses are
often no more than a one-dimensional set of hazarded guesses. They fail to realize that
risks are multi-dimensional so that one risk can be dependent on and a cause of another.
A well-known example from the literature is Feynman’s analysis of the Challenger acci-
dent. NASA and Morton Thiokol grossly undercalculated the risk probabilities for the
different components of the space shuttle and its operation. Feynman calculated that the
probability would be 1 in 50 that the operation would fail. Later, when the Columbia
accident happened, he was sadly vindicated. 2 in 135 missions failed. Feynman’s app-
roach and analysis were, in our view, much in line with FFA and abductive logic [32]. We
only sampled a few project participants (n = 4) to identify their scientific standings and
they diverted. On scrutinising some follow up interviews we found indications implying
that the positivist and non-positivist divide existed, and that there was a lack of dynamics
in the change between different standings in different relevant situations. We will hold
this in abeyance, but our second hypothesis seems to be confirmed.

4.3 Port Research Projects

The research projects studied were typical triple helix projects initiated by the govern-
ment or a research institution and financed by national or interregional governments. The
42 A. M. Christiansen and K. Vidskjold

objective of the projects was to develop sustainable ports for the future and the scope
was to develop and implement different types of autonomy and automation for the port
operational processes, technology for electrification of port equipment and shore power.
Different systems for digitalisation of both the operational processes and several port
business processes were also parts of the scope of these projects. The ports participating
in the research projects are Norwegian and Danish small and medium- sized commercial
ports with a local or regional operational scope. They were handling a mix of different
types of cargo spanning containers, dry bulk and general cargo. The ports were typically
equipped with “ship to shore” gantry cranes or Mobile harbour cranes for loading and
unloading ships, and reach stackers, straddle carriers and forklifts for internal cargo
handling. In total, the potential for automation, autonomy and digitalization was high.
Nevertheless, the projects did not deliver much within their duration and afterwards. The
potential for electrification was also high, but here the project deliverables were much
better, and could with some reserve, be considered much more successful. Prominent
drivers identified within these projects were market, public opinion and sustainability,
and barriers were governance, public opinion and policy. Public opinion was a driver
in some ports in some situations and a barrier in others. To further detail this would be
impossible without compromising the anonymity of the three ports investigated.
Sustainable Public Transportation – The “City Ferry” Project. Several decades
ago, local ferries were a common mode of travel, but development of vehicles and
road infrastructure made the local ferries obsolete. Currently, there are some cities and
municipalities in Norway who still operate local ferries for public transportation. Due to
traffic congestion and the focus on a sustainable development of the society, there is an
increasing trend among municipalities to consider alternative transportation, including
use of waterways. We have participated in various projects aimed at establishing envi-
ronmentally friendly ferries as part of the public transport system. The research projects
were typical triple helix projects which had been initiated by a local municipality or a
private company, usually a property developer. The projects were financed by private
stakeholders, municipalities and national government through research initiatives.
The city ferry projects started in 2019. Initiated by a group of local politicians, the
project was established with the USN and actors from the industry. The project objective
was to develop a concept and launch a city ferry for testing. USN conducted a survey
showing that the public was positive about using a ferry for transportation on a frequent
basis. The industry partner presented different concepts and demonstrated a feasible
ferry solution. The project report stated that the public was positive, the technology was
readily available and the infrastructure was in place.
The project results and deliverables were also presented in public meetings of local
politicians and the local committee of transportation. During the spring of 2019, the
local council concluded that they did not see the need for a ferry and there was therefore
no need for further appraisal. Academia and to some extent the public did not agree with
this conclusion. Prominent drivers identified here were technology and sustainability,
and barriers were governance and policy.
Ship X - The Autonomous Ships Project. There is an increasing interest in developing
autonomous ships among different stakeholders in the maritime transportation industry.
The stakeholders are challenging the limits in terms of what is feasible within the existing
 Management and Emerging Technology in Maritime Logistics 43

laws and regulations and public acceptance of unmanned transportation modes. A project
“ship X” was started in 2016 and the goal was to develop “the world’s first fully electric
and autonomous ship”. The ship was launched in 2022. It is fully electric driven but is
awaiting governmental approval to be fully autonomous. In short, this project proved to
be somewhat successful and achieved positive media attention regarding autonomous
ships. The project execution was challenging, mostly due to many delays in deliverables
and numerous safety and legislative issues that had to be resolved. All drivers identified
as major in the ship X project were the same as those in the seaport and city ferry
projects. However, the analysis of this project showed a definite change towards a more
non-positivist approach. “Truths” went from being non-negotiable to being discussed,
negotiated and changed. In particular, rules and regulations revealed a more dynamic
approach towards change during the project. Prominent drivers identified here were
technology and sustainability, and barriers were governance and policy.

4.4 Common Drivers and Barriers Identified in all Projects

In Table 4 we have compiled a more detailed matrix of forces based on the seven main
forces identified in Table 2. The connections are sometimes complex, and rather one-to-
many than a one-to-one relationship. As an example, risk aversion, which can be a strong
barrier to change and innovation [33], is embedded in all entries under the government
column. In moving from the seven major forces to the elaboration in Table 4, we used
FFA and abductive logic.

Table 4. Major project development forces identified

Driving Axis Public Government Research Industry

Safety Trust HSE Legislation New Technology Values
Trends ISM code Improve Training Profit
Values Incentives Strategy
Vision
Security Trust ISPS Code Surveillance Viability Study
National Legislation Technology
Sustainability Values HSE Legislation Eco Technology Values
Trends Incentives Green Energy Long-term Profit
Feasibility Comfort National Legislation Qualify Technology Feasibility Study
Mobility Training and TRL, TRI
Assessment
Viability Trends Incentives Improve Practice Viability Study
Cost/Benefit Policies Improve Efficiency Business Models
44 A. M. Christiansen and K. Vidskjold

5 Discussion
There are two parts to this discussion. Firstly, we discuss the implications of the identified
theoretical perspectives among different project members and stakeholders. Secondly,
we discuss the findings of the FFA regarding barriers and drivers concerning successful
project management. This article started with the perils of achieving a successful out-
come of projects dealing with both technological and societal changes and we end our
discussion by returning to the initial problem.

5.1 Positivism versus Non-positivism

We started this project by addressing the issues and challenges in R&D projects dealing
with both emerging technologies and humans. This was a challenge because the different
standings in the natural and social sciences and a lack of common ground. By analysing
projects dealing with identified drivers that enable or restrain change and unidentified
drivers that do the same, we have chosen a more holistic approach to project analysis and
evaluation. Together with all the drivers, these form connections and interdependencies
within the projects. Drivers that are left unidentified and or unknown can create restrain-
ing internal forces within any project and cause failure to achieve desired change. All
data we gathered showed that it was important to identify and address all major drivers
within a project. This included those that seemed less important at project initialisa-
tion. The positivist approach would say that what is not observable is not there, but
the non-positivist approach would be that forces can emerge and change during project
execution.
Towards a Common Philosophical Perspective? The “Ship X” project can introduce
an important and vital shift towards a more common philosophical perspective among
industry, government and academia. The ship has contributed to materialising the con-
cept autonomous technology, a concept which until recently has existed as drafts and
hypothetical discussions. The “Ship X” project enabled stakeholders to explore the bene-
fits of the technology and how to utilise the technology to benefit from more efficient and
sustainable shipping. Further, this project has developed a broader understanding regard-
ing human interaction with autonomous technology and the spawning of many projects
in various areas of electric and autonomous ship operations. The two autonomous sea
drones currently plying the Oslo Fjord are one instance of many examples.
Balancing the Gap Between Human- and Technology Centred Ship Operations. In
close cooperation between academia and regulatory bodies, the concept of remote oper-
ational center [ROC] has been developed. The control centre requires personnel onshore,
and initially also on the ship, to collaborate on exploring and developing new knowl-
edge and practice, including human factors and technology, in common operations. An
autonomous ship is not an “unmanned ship” but an “automated transportation mode”.
Some of the reasons for this common misunderstanding stem from different philosophi-
cal perspectives or standings among the project participants. From an industry perspec-
tive, an autonomous ship is just an expansion or change of an already existing automated
supply chain. From a governmental perspective, an autonomous ship is a ship without
people and representing something radically different.
 Management and Emerging Technology in Maritime Logistics 45

Many of the companies developing autonomous ships are not traditional shipown-
ers, but they are companies seeking more effective and environmental alternatives for
moving their cargoes. The technology behind ships like the “ship X” has existed since
the early 1980s. Dynamic Positioning (DP) was developed and implemented in the mar-
itime industry during this period and other automation technologies have emerged in
the following decades, making ship-operations less dependent on human intervention.
Technology is one of the main driving forces in the shipping industry and is based on
a philosophy that automated technology with less human interaction can make ship-
ping safer and more effective and ultimately sustainable. In the same period, rules and
regulations governing the shipping industry have verged towards supporting the auto-
mated development of ship operations, but the philosophy remains unchanged. Ship
operations are still human centred, regarding who is controlling the ship and rules and
regulations are based on this human-centred philosophy. One informant told us that The
International Maritime Organisation (IMO) was already in the 1980s working on how to
implement autonomous shipping. Several states and their governmental bodies have in
the same period worked on how to approach autonomous technology, but the work has
mainly focused on the human factor. The technology supporting autonomous shipping
is relatively new but still considered as emerging and not fully supported by rules and
legislation. Likewise, the political discussion and governmental approach to autonomy.
But they are still working from different scientific perspectives. The “Ship X” case shows
that there could be two different philosophical perspective that could explain the mis-
match between industrial development and governmental regulation supporting industry.
The industry is developing “automated transportation modes”, and the government are
developing regulations supporting “unmanned ships”.

5.2 A New Kind of Project Management?

Government Participation in the Development Phase. The Norwegian Maritime

Authority has established a new section for new maritime technology. This section will
function as an adviser and collaborate with the shipowners in the development process.
This is a more dynamic approach towards the possible approval and implementation
of new technologies in the shipping industry. The traditional process for approval has
been that the shipowner, together with the shipbuilder, develop new concepts within the
existing regulatory framework before applying for approval from the authorities. The
traditional approaches are representing higher risks regarding approval of new concepts.
Traditional approval processes create a barrier between government and eurocrats and the
industry. Some authors advocate for a more reflexive governance in projects like these:
“…by (1) recognize the politics of learning for sustainable development and develop
safeguards against domination and capture by powerful actors, and (2) systematically
consider the embedding of governance designs in political contexts and their ongoing
dynamics for political fit” [34]. This could contribute to lower risk and establishing a
more common and universal set of strategies in R&D and innovation projects. More-
over, this would facilitate a more agile set of regulations and facilitate the management
of projects focused on emerging technology.
46 A. M. Christiansen and K. Vidskjold

Managing Forces. The city ferry project failed to deliver the desired goals. Even though
several central forces acted as drivers for the project to succeed, most of the identified
forces in this project were acting as drivers for sustainable public transportation. Such
is the trend towards a more sustainable society, namely, public opinion was positive to
ferries and the municipality had sustainable mobility as one of the key objectives in their
strategy for transportation. The main barrier to the success of this project appeared to
be the high degree of uncertainty surrounding the ownership of a ferry that would take
the risk of investing and operating a ferry. The municipality had no previous experience
of operating a ferry. Based on the degree of uncertainty involved with a ferry, it was
less risky for the municipality to reject further investigation of the possibilities for a city
ferry. Risk aversion was identified as a strong force in the projects identified. Mostly as
a restraining force, but in our view with the potential to be inverted or neutralised.
This case shows that, despite all good intentions and several key drivers working
towards a desired outcome, this was still not enough to overcome the degree of uncer-
tainty in developing alternative transportation. The municipality made their decision
based on what was most predictable and involved less risk for the municipality. Using
Lewin’s force field analysis in the beginning of this project could have identified this
barrier and enabled neutralising some of the uncertainty and maybe changing the out-
come of the project in line with the intentional goals. Both our hypotheses were at least
partially confirmed. Few traces of the restraining forces identified in our analysis were
identified in the initial project planning framework, and if they were, their effects were
underestimated in what we could identify as project risk analysis. This was especially
relevant regarding the resistance these projects met from bodies upholding rules and
regulations. We identified these forces as governance in all investigated projects. We
hold this as a solid confirmation of our first hypothesis.

6 Conclusion
This paper shows that Lewin Force Field Analysis could be a potentially strong tool
for identifying and understanding how the different driving and restraining forces are
impacting the outcome of R&D projects. This paper provides the project management
with a broader understanding and a wider range of tool for planning and executing
projects dealing with human effort and emerging technology in a common system. This
paper focused on identifying and describing different forces in these projects. However,
more work is required to comprehend how to address these forces and utilise them to
the benefit of the project and the pertinence of such findings to other projects in other
settings than those investigated herein.

References
1. Dooms, M.: Stakeholder Management for Port Sustainability: Moving From Ad-Hoc to
Structural Approaches, pp. 63–84. Elsevier, Green Ports (2019)
2. Ishizaka, A., Labib, A.: Analytic hierarchy process and expert choice: benefits and limitations.
Or Insight 22, 201–220 (2009)
 Management and Emerging Technology in Maritime Logistics 47

3. Munier, N., Hontoria, E.: Uses and Limitations of the AHP Method. Springer, Cham (2021).
https://doi.org/10.1007/978-3-030-60392-2
4. Kvale, S.: Validation as Communication and Action: On the Social Construction of Validity
(1994)
5. Rorty, R.: Philosophy and Social Hope. Penguin UK (1999)
6. Flyvbjerg, B., Bruzelius, N., Rothengatter, W.: Megaprojects and Risk: An Anatomy of
Ambition. Cambridge University Press, Cambridge (2003)
7. Epstein, B.: The Ant Trap: Rebuilding the Foundations of the Social Sciences. Oxford Studies
in Philosophy o (2015)
8. Lewin, K.: Principles of Topological Psychology. Read Books Ltd., Redditch (2013)
9. Swanson, D.J., Creed, A.S.: Sharpening the focus of force field analysis. J. Chang. Manag.
14, 28–47 (2014). https://doi.org/10.1080/14697017.2013.788052
10. Adelman, C.: Kurt Lewin and the origins of action research. Educ. Act. Res. 1, 7–24 (1993)
11. Eisenhardt, K.M.: Better stories and better constructs: the case for rigor and comparative
logic. Acad. Manag. Rev. 16, 620–627 (1991)
12. Eisenhardt, K.M.: Building theories from case study research. Acad. Manag. Rev. 14, 532–550
(1989)
13. Lewin, K.: The solution of a chronic conflict in industry. In: Proceedings of the Second Brief
Psychotherapy Council, pp. 3–17 (1944)
14. Lewin, K.: Defining the “field at a given time.” Psychol. Rev. 50, 292–310 (1943). https://
doi.org/10.1037/h0062738
15. Lewin, K.: Resolving Social Conflicts And Field Theory In Social Science. American
Psychological Association, Washington, DC (1997). https://doi.org/10.1037/10269-000
16. Lewin, K.: Resolving Social Conflicts. Harper International, Buffalo (1948)
17. Weigend, G.G.: Some elements in the study of port geography. Geogr. Rev. 48, 185 (1958).
https://doi.org/10.2307/212130
18. UNCTAD: Review of Maritime Transport. UN, S.l. (2020)
19. Paddeu, D., Calvert, T., Clark, B., Parkhurst, G.: New technology and automation in freight
transport and handling systems, p. 71 (2019)
20. Jovic, M., Kavran, N., Aksentijevic, S., Tijan, E.: The transition of croatian seaports into smart
ports. In: 2019 42nd International Convention on Information and Communication Tech-
nology, Electronics and Microelectronics (MIPRO), pp. 1386–1390. IEEE, Opatija (2019).
https://doi.org/10.23919/MIPRO.2019.8757111
21. Sorgenfrei, J.: Port Business. Walter de Gruyter GmbH & Co KG, Berlin (2018)
22. Johnson, R.B., Onwuegbuzie, A.J.: Mixed methods research: a research paradigm whose time
has come. Educ. Res. 33, 14–26 (2004)
23. Lincoln, Y.S., Guba, E.G.: Naturalistic Inquiry, vol. 75. Sage, Upper Saddle River (1985)
24. Crotty, M.: The Foundations of Social Research: Meaning and Perspective in the Research
Process. Sage, Upper Saddle River (1998)
25. Eikeland, O.: Action research and organisational learning: a Norwegian approach to doing
action research in complex organisations. Educ. Act. Res. 20, 267–290 (2012). https://doi.
org/10.1080/09650792.2012.676303
26. Ågerup, M.: Velfærd i det 21. århundrede. People’s Press (2017)
27. Leydesdorff, L., Etzkowitz, H.: Can ‘the public’be considered as a fourth helix in university-
industry-government relations? In: Report on the Fourth Triple Helix Conference. Science
and Public Policy, vol. 30, pp. 55–61 (2002)
28. Leydesdorff, L., Etzkowitz, H.: Emergence of a Triple Helix of university—industry—gov-
ernment relations. Sci. Publ. Policy 23, 279–286 (1996)
29. Etzkowitz, H.: The Triple Helix of University - Industry – Government, vol. 18 (n.d.)
30. Epstein, B.: The Ant Trap: Rebuilding the Foundations of the Social Sciences. Oxford
University, Oxford (2015). https://doi.org/10.1093/acprof:oso/9780199381104.001.0001
48 A. M. Christiansen and K. Vidskjold

31. Lewin, K.: Field Theory in the Social Science. Harper Brothers Publishers, New York (1951)
32. Feynman, R.P.: “What Do You Care What Other People Think?”: Further Adventures of a
Curious Character. W. W. Norton & Company, New York (2011)
33. Darsø, L.: Innovationspædagogik: kunsten at fremelske innovationskompetence. Samfund-
slitteratur, Frederiksberg (2011)
34. Voß, J.-P., Bornemann, B.: The politics of reflexive governance: challenges for designing
adaptive management and transition management. E&S 16, Article no. 9 (2011)
 Streamlining the Execution of Maritime
Commissioning with a Digital Assistance
System

Tim Maximilian Jansen(B) , Oliver Karl, Ahmed Elzalabany,

and Hermann Lödding

Hamburg University of Technology, 21073 Hamburg, Germany

[email protected]

Abstract. Providing the best possible support to commissioning engi-

neers is critical for efficient execution of commissioning and on-time deliv-
ery of a shipbuilding project. This paper contributes to current research
on the digitization of maritime commissioning and builds on existing
approaches to a digital twin. Based on an analysis of the execution pro-
cess, this paper presents a human-centric digital assistance system that
is designed to provide the commissioning engineers with an improved
information supply and to increase situational awareness in the com-
plex and error-prone ETO process. In addition, the system can be con-
nected to the ship automation system via OPC UA to display, check,
and document sensor values directly in the assistance system. Problems
and rework items can be documented context-sensitively and with little
effort and fed back into the digital twin. The automated documenta-
tion and continuous enrichment of the digital twin help reduce effort,
make the commissioning process more transparent, and progress indica-
tions more resilient. The prototype implementation at three shipyards
promises improvements in productivity and quality.

Keywords: ship commissioning · digital twin · human-centric

production systems · digital assistance system

1 Introduction
The shipbuilding industry faces economic pressure and strives for more efficient
solutions to secure economically stable operations [1]. However, shipbuilding
is a complex Engineer-to-Order (ETO) process that requires high engineering
effort [2] and is often one-of-a-kind [1,2]. ETO manufacturing is generally more
complex and dynamic [3], but solutions to increase efficiency remain scarce [1].
Additionally, shipbuilding lags behind other industries in terms of digitization,
despite the need for integrated IT solutions to handle the complexity, uncer-
tainty, and dynamic shipbuilding processes [1].
One critical phase for shipbuilding is commissioning, which involves ensuring
that a ship is ready to operate and satisfies the classification and customer
c IFIP International Federation for Information Processing 2023
Published by Springer Nature Switzerland AG 2023
E. Alfnes et al. (Eds.): APMS 2023, IFIP AICT 691, pp. 49–63, 2023.
https://doi.org/10.1007/978-3-031-43670-3_4
50 T. M. Jansen et al.

requirements by performing inspections, tests, and trials [4]. Commissioning is

part of a thorough quality assurance process [5]. It is also of financial relevance,
as the handover to the customer ultimately ensures the shipyard’s cash flow.
However, commissioning lacks end-to-end digitization and human-centric
information provision, highlighting the generally low level of digitization in the
shipbuilding industry [1]. In addition, shipbuilding tasks rely heavily on work
instructions and design documents, which are often still in paper form, limit-
ing access to up-to-date information and resulting in more time spent gathering
information. Alongside this, there is a high level of uncertainty due to disrup-
tions in the ETO process. Furthermore, processes are difficult to track, leading to
delays because of the false impression that the project is on time [4]. These delays
can have a direct impact on the customer and customer satisfaction, stressing
the need for improvements.
Appropriate integration between IT systems down to the production level has
yet to be achieved [1]. Therefore, there is a need for research to develop more
efficient and integrated IT solutions that provide up-to-date and high-quality
information, particularly for the commissioning process. This paper builds on
existing approaches to digitizing the maritime commissioning process and com-
plements them with a human-centric digital assistance system for commissioning
execution.
In the following, the current state of research on maritime commissioning
is presented, and the process of commissioning execution is analyzed in more
detail (Sect. 2). Based on the deficits identified, a digital assistance system for
the execution of maritime commissioning is presented (Sect. 3), and the associ-
ated potentials and limitations are described (Sect. 4) before an outlook is given
(Sect. 5).

2 Maritime Commissioning
This chapter introduces the basics of commissioning and digitization approaches.
The commissioning process is then analyzed in more detail, with a focus on the
execution phase, and deficits are identified.

2.1 Current State of Research

The following section introduces the fundamentals of commissioning, digital

assistance systems, and the use of digital solutions in commissioning execution.

Commissioning. The term commissioning originates from shipbuilding and

means going through an extensive quality assurance process for systems, com-
ponents, and crew [5]. Other industries have taken the idea and implemented
similar concepts. For example, the commissioning of industrial facilities, such
as those in the gas and chemical industries, is a discipline that involves test-
ing, calibration, and verification of all systems and components to ensure proper
 Digital Maritime Commissioning Assistant 51

operation of the facility [6]. In building construction, commissioning is often per-

formed as a quality assurance process to ensure the intended functionality of all
equipment and systems, and an efficient operation [5]. Today, building commis-
sioning also serves as a powerful lever for reducing energy consumption, costs,
and greenhouse emissions [7].
Maritime commissioning is a complex and critical process that involves ensur-
ing that a ship’s systems and equipment are functioning properly and safely
before the ship is put into operation [4]. The commissioning process typically
includes testing, verifying, and documenting the performance of various systems
and subsystems, such as propulsion, electrical, and navigation systems.

Digital Assistance Systems. Digital assistance systems support people in

capturing and processing information as well as performing work by provid-
ing selective and context-sensitive information [8,9]. The main goals of digital
assistance systems are to increase productivity, reduce training times, search
efforts, and execution errors, and provide the best possible support for people
in complex tasks [10]. In production, digital assistance systems are increasingly
used to support workers in manufacturing and assembly activities. The aim is to
support or improve the existing skills of the workers without replacing or over-
ruling them [11]. Keller et al. categorize typical functions of digital assistance
systems according to sensory and cognitive assistance [12]. Sensory assistance
includes the documentation of machine, product, and process data as well as
their monitoring. Cognitive assistance, in turn, comprises the provision of infor-
mation in textual or visual form as well as the provision of learning content. In
addition, support in decision-making or context-sensitive communication with
responsible individuals is also a component of cognitive assistance [12]. Frequent
areas of application in the production environment are, for example, interactive
instructions in assembly or maintenance and servicing [8]. In this context, mobile
devices such as tablets and smartphones, but also smartwatches and AR glasses,
are mostly used as display and interaction devices. Digital assistance systems
can only realize their full potential when they are embedded in the IT infras-
tructure of production and can thus also react to situations with information or
instructions, e.g., based on sensor data [8,10].
Interviews with project managers from three major shipyards in Germany
have shown that the commissioning process is demanding a holistic digitization
approach. Some aspects, such as the use of digital checklists to document test
results or the use of digital punch lists to track necessary rework and quality
issues, have already been addressed by some shipyards. However, there is no
comprehensive digitization of the commissioning process in the maritime indus-
try, which could potentially lead to inefficiencies, delays, and errors.

Digital Solutions for Commissioning. Several industries have developed

software solutions for commissioning processes that focus on documenting and
tracking the performance of facilities and systems. For example, CxPlanner is a
software solution that focuses on ASHRAE and LEED (Leadership in Energy
52 T. M. Jansen et al.

and Environmental Design) requirements for building commissioning [13]. The

software helps provide an efficient, safe, and healthy facility by documenting
the facility’s performance [14]. Similarly, software solutions have been developed
for commissioning in the oil, gas, petrol, and pharmaceutical industries, which
often incorporate industry-specific standards and therefore cannot be applied to
commissioning in maritime one-off manufacturing [15]. Commissioning software
solutions typically include several functions and features that help streamline the
commissioning process and ensure that all necessary tests are completed and doc-
umented. These functions include activity overviews, test checklists, locations on
2D drawings, punch and issue lists, progress indication, and report creation. The
activity overview function provides an overview of the commissioning process and
its progress. The test checklist function enables the creation and management
of test checklists and helps ensure that all necessary tests are performed. The
automatic location on 2D drawings allows for tracking the location of various
equipment and systems on drawings. The punch and issue list function tracks
any necessary rework or quality issues that arise during commissioning. The
progress indication function provides real-time progress updates on the commis-
sioning process. Finally, the report creation function enables the creation and
distribution of commissioning reports to relevant stakeholders. [13,16]
Approaches for maritime commissioning software also exist. Elzalabany et
al. [17] describe a digital twin for the commissioning process and propose a
template-based approach for creating commissioning test specifications. Elzala-
bany et al. also present an AR-based commissioning application for tablets. It
enables locating the test area and identifying the system and components to be
tested. The application is also designed to increase the situational awareness of
commissioning personnel by providing up-to-date information about currently
blocked areas or zones on the ship. Finally, it allows test data, such as engineer-
ing measurements and photographs, to be fed back into the digital twin.

2.2 Commissioning Process

A shipbuilding project starts with the design and development phase. This is
followed by manufacturing and assembly. Commissioning is a phase between the
production and operation of a ship. It runs partly parallel to production and
concludes with classification and customer handover. This is followed by the
operation and decommissioning of the ship (Fig. 1) [4,18].
Elzalabany et al. [19] subdivide the commissioning phase into three phases:
authoring, scheduling, and execution.
The authoring process includes the compilation of all information necessary
for the commissioning tests and the subsequent creation of the test specifications.
In addition, the functional dependencies between the tests are also defined here.
In the scheduling phase, the tests are sequenced and scheduled, taking into
account the previously defined dependencies, available resources, and other sche-
duling constraints.
During the execution phase, the necessary documents and test tools for the
tests are compiled before they are carried out by the commissioning personnel.
This is followed by the documentation of the test execution.
 Digital Maritime Commissioning Assistant 53

Fig. 1. Shipbuilding and commissioning phases (adapted from [18, 19])

2.3 Commissioning Execution

This paper focuses on a digital assistance system for the execution process,
which is why this process is covered in more detail. The commissioning pro-
cesses of three major German shipyards, each with more than 250 employees,
were mapped through expert interviews and workshops with project managers
from the shipyards. In addition, commissioning documents were reviewed and
analyzed. The shipyard-specific processes were then used to derive a generalized
process for commissioning execution, which was reconfirmed by the previously
interviewed experts.
The commissioning execution process can begin as soon as a commissioning
test has been released. The process is divided into three stages: preparation,
execution of tests and follow-up (Fig. 2).

Fig. 2. Commissioning execution stages

54 T. M. Jansen et al.

In the preparation stage, the foundations are laid for the further course of
commissioning, and it is ensured that the necessary procedures, documents, and
equipment are in place. The required documents and test records are collected
and checked for completeness. The documents can be in paper form or digitally
supported. The next step is familiarization with the test documents, during
which ambiguities are eliminated using documents, drawings, and, if necessary,
3D models. Also, crucial at this stage is checking that the prerequisites are met,
which involves a manual inspection of the relevant areas and systems. In addition,
the required test equipment and tools are gathered, and navigation to the test
location is performed. Identification of the system or component to be tested
is also performed, and a visual inspection of the completeness of the system is
carried out. Only when all these steps have been successfully completed can the
actual tests be performed.
The execution of tests involves working through the checklist, which can be
paper-based or digitally supported. If values have to be checked, such as mea-
surements, signals, etc., the reference values are contained in the test protocol
or have to be extracted from the test documents.
The follow-up stage involves documenting deficiencies and other issues, which
are compiled into an issue or punch list. The test documentation is then finalized,
and if paper-based, it is scanned for archiving purposes.
It is clear that preparation is critical to successful commissioning testing and
is extensive due to the complex interdependencies.

2.4 Process Deficits

The commissioning of ships is a critical aspect of ensuring their proper function
and safety. Additionally, this process has a direct impact on the customer, both
in terms of on-time delivery and quality. However, the current process suffers
from several deficiencies that can compromise its effectiveness and efficiency.

Difficulty in Accessing Dependency Information. A major problem is the

difficult and time-consuming consideration of dependencies. The current proce-
dure does not automatically assess whether all prerequisites, such as previous
tests, have been met. This is currently done manually, which can lead to delays
in commissioning.

Manual Provision of Information. Another problem is the inadequate inte-

gration and manual provision of information. For example, the installation sta-
tus of systems and components is not automatically forwarded from the previous
lifecycle phase. In addition, there is no up-to-date information on the availabil-
ity of documents and resources, and documents and test records are collected
manually.
Further, there is no current information about restricted areas on the ship,
which makes navigation to the test site difficult and could ultimately result in
the test site being inaccessible to commissioning personnel.
 Digital Maritime Commissioning Assistant 55

Lack of Integration with Ship Automation. There is a lack of integration

with ship automation to obtain real-time system and signal data. This lack can
make it difficult to collect and verify measurements, leading to manual efforts.

Lack of Automated Documentation. Documentation is crucial for prov-

ing compliance with classification and customer requirements. However, it still
involves manual effort, such as filling out inspection logs and punch lists or scan-
ning for archiving purposes. Paper-based logs are also not suitable for reuse in
subsequent lifecycle phases, such as operations. In addition, handwritten logs
may not be legible, which can lead to errors during commissioning.

3 Digital Commissioning Assistance System

This chapter outlines the concept and implementation of the digital assistance
system for maritime commissioning. First, an existing solution for creating com-
missioning tasks and a digital twin for commissioning are introduced. Then,
the necessary system architecture is explained. Finally, the development of the
digital assistance system for performing commissioning is described, and the
functions and user interfaces are presented.

3.1 Commissioning Authoring and Digital Twin

As depicted in Fig. 1, authoring is the first stage of the commissioning process.
Thus, to initiate the commissioning execution process, the commissioning exe-
cution assistant must have access to a system capable of authoring and storing
all the necessary data for maritime commissioning. For this purpose, the com-
missioning execution assistant presented in this paper utilizes an existing digital
authoring system that provides a digital twin infrastructure for maritime com-
missioning [19].
The digital twin is connected to the shipyard’s Product Lifecycle Manage-
ment (PLM) system to obtain essential data, such as the bill of material (BOM),
assembly groups, and technical specifications. The digital authoring system uses
this data to generate commissioning tests, including details for execution such
as location information, resources, and documents.
The authoring system adopts a template-based authoring workflow that
allows the user to create a generic template for each component class in the ship.
The template encompasses steps for component testing and technical parame-
ters, defining the types of readings to be captured during the testing procedure.
These templates are generic enough to be used across different ships. The digi-
tal twin offers a matching algorithm that analyzes the data retrieved from the
PLM system and automatically generates commissioning tests that adhere to the
specifications defined in each template. This approach can save time compared
to the traditional workflow of creating every commissioning test from scratch.
Furthermore, the digital twin has the ability to arrange the created tests
dynamically into a graph of technical dependencies based on rules established
by the user and status information obtained from the production process. This
feature is crucial for achieving a practical execution sequence in real time.
56 T. M. Jansen et al.

3.2 System Architecture

As already outlined in Chap. 1, maritime commissioning and shipbuilding in
general demand end-to-end digitization from higher-level IT systems to the exe-
cuting commissioning engineer in the production area at the shipyard, as well
as consolidation of information and automated workflow concepts. In order to
provide the commissioning engineer with the best possible supply of information,
it is therefore necessary to connect several IT systems and databases.
The system architecture for the digital assistance system for maritime com-
missioning execution is shown schematically in Fig. 3.

Fig. 3. System architecture for ship commissioning assistance system

The digital twin platform represents the central system for the digital com-
missioning process. On the one hand, this contains all the data on the commis-
sioning tests from the authoring and scheduling phases, which are combined in
the digital twin. On the other hand, information from other IT systems is also
bundled here and processed for the digital assistance system. There is a bidirec-
tional connection between the digital twin platform and the digital assistance
system so that relevant test information can be provided to the commission-
ing engineer and, at the same time, information such as completed test steps,
recorded measured values, and remaining work points can be fed back into the
digital twin.
Regarding the production IT landscape, connection to higher-level IT sys-
tems, such as enterprise resource planning (ERP) and product lifecycle man-
agement (PLM) or product data management (PDM) systems, is also required.
From this, information is extracted on the availability of employees, test tools,
 Digital Maritime Commissioning Assistant 57

and other required documents. In addition, there is a connection to the invoice

directory and the 3D visualizations of the systems and components. The instal-
lation status of the components can also be queried via this interface.
The connection of the ship’s automation to the digital twin enables the uti-
lization of sensor and signal data in real time via the digital assistance system.
The data exchange takes place via OPC UA. This standard is also being used
more and more in ship automation [20].

3.3 Commissioning Execution Assistant

Elzalabany et al. [17] describe an approach for a tablet-based assistance system
for performing commissioning. It includes navigation assistance to the testing site
through 2D and 3D visualization and the use of AR. Furthermore, it describes
the possibility of manually recording measured values and transferring them back
to the digital twin. The use of IoT technologies for more efficient measurement
recording has already been considered [19].
The described approach was further developed and transferred into a web
application to meet the challenges of the maritime commissioning process. This
has the advantage that the digital assistance system is not tied to a specific
hardware, but can be called up via any browser.
Firstly, a requirements analysis was conducted based on the identified defi-
ciencies of the commissioning process. During the development phase, the func-
tionalities were successively provided to three shipyards for testing with real
data. The experts at the shipyards gathered internal feedback from end users,
particularly for the user interfaces. This feedback was collected in weekly meet-
ings and regular workshops, and the functionalities were adapted accordingly, or
new functionalities were added. The close exchange with three different shipyards
is intended to ensure a high degree of applicability of the concept throughout
the shipbuilding industry. The assistance system was implemented as a proto-
type, as shown in Fig. 3. The following functionalities and aspects of the digital
assistance system are described in more detail:
– user interface
– information supply for improved situation awareness
– navigation and identification of systems and components
– checklists for commissioning tests
– connectivity to ship automation
– documentation of rework items and issues

User Interface and Situation Awareness. The user interface is generally

divided into three areas (Fig. 4). On the left side, there is an overview list of the
tests or test steps. Depending on the view, the 2D plan of the ship with locations
of the tests or the 3D model of the current system is displayed in the upper right-
hand area. The lower area contains further information on the current test step.
The application is responsive, so that the layout automatically adapts to the
screen of the display device and thus provides the best possible display for the
commissioning engineer.
58 T. M. Jansen et al.

Fig. 4. Commissioning assistance system showing the released commissioning tests

(left) and their location on the 2D map (right)

The test overview shows the released commissioning tests and their status.
To increase the situational awareness of the commissioning engineer and to avoid
unnecessary work steps, there are further indicators (Fig. 4).
The installation status shows whether the systems or components to be
tested have already been completely installed on the ship. The dependency sta-
tus provides information about dependent tests and whether they have already
been completed. For example, the signal test of the engine cannot yet take
place because the upstream test, the electricity test, has not yet been com-
pleted. Finally, information about the availability of the resources and documents
required for the test is provided by an availability indicator. The indicators are
represented by the signal colors red and green, so that the commissioning engi-
neer can see at a glance whether a test is feasible or not.
As with Elzalabany’s approach, there is a 2D map of the ship with markers
of the tests to help navigate to the test locations (Fig. 4). Blocked areas on the
ship are also displayed on the map. Once a test has been selected, it is possible
to switch to a representation of the 3D model that highlights the system or
component being tested (Fig. 5). This allows for easier identification, especially
when several similar components are installed.
 Digital Maritime Commissioning Assistant 59

Fig. 5. Commissioning assistance system showing the steps of the current commission-
ing test (left), the highlighted test component in the 3D model (upper right) and the
current measurement values via OPC UA (lower right)

Checklists and Connectivity to Ship Automation. Figure 5 shows the

display when a concrete test has been selected. The overview of the test steps
on the left side shows the work instructions and the status of the test step. In
the detail area of the test step on the bottom right, further information such as
linked documents, drawings, and photos already taken are provided.
In the digital twin, the commissioning checks are linked to components and
their signals. When a commissioning test is called up, this information is used to
automatically establish a connection to the corresponding OPC UA server, and
the corresponding sensor values are displayed in the digital assistance system
(Fig. 5). If reference values for this signal are stored at the step, an automatic
check is performed to determine whether the current value corresponds to the
specifications. This is indicated by a colored marker. The current sensor value
can be accepted and scanned with a click.
60 T. M. Jansen et al.

Fig. 6. Commissioning assistance system showing the creation of an issue with prefilled
component and location information

Documentation of Rework Items and Related Issues. If residual work or

other problems occur during the commissioning test, they can be added to the
issue list directly in the context of the test or test step (Fig. 6). In doing so, the
entry is automatically enriched with information about the test, the system to
be tested, and the location.

4 Validation and Limitations

The aim of the digital assistance system for performing maritime commission-
ing is to provide the commissioning engineers with the best possible, up-to-date
information so that they can perform their work better and make more informed
decisions. At the same time, this is also intended to increase productivity in mar-
itime commissioning. The digital assistance system holds the following potentials.
The use of the digital twin eliminates the need for manual compilation of
documents relevant to the inspection, such as inspection records and drawings.
The inspections in the digital twin contain references to the documents that can
 Digital Maritime Commissioning Assistant 61

be retrieved via the interface to the PLM or PDM system. Moreover, this way
of working ensures access to the most up-to-date documents. The installation,
dependency, and resource status indicators are designed to improve the com-
missioning engineer’s situational awareness and eliminate the need for manual
checking of preconditions. This makes the feasibility of a test directly apparent
and reduces the effort required to check the preconditions.
Dynamic display of tests on 2D maps and in the 3D model allows interactive
and faster navigation to the test location. The display of blocked areas can also
improve the commissioning engineer’s situational awareness.
The automatic transfer of signal values via OPC UA enables faster processing
of the test steps as all information is available in the same digital assistance
system. Since there are no media breaks, the documentation effort is also lower,
and the documentation is less prone to errors.
By recording issues in the context of the commissioning test, the documenta-
tion effort is reduced, and media breaks are avoided. In addition, they are easier
to retrieve afterward. The data consistency in interaction with the digital twin
allows a more informed and detailed statement about the process progress of
the commissioning, so that possible false perceptions of the process progress are
reduced and late detected triggers for delays are prevented.
A first validation of the concept was possible through the prototype imple-
mentation of the digital assistance system and its test use on three shipyards
with real data from different systems, including electrical and mechanical engi-
neering components. In order to verify the previously described potentials of the
digital assistance system with regard to productivity and quality of the maritime
commissioning process, tests and evaluations are currently being carried out at
three shipyards. The usability of the digital assistance system will be determined
primarily through productivity measurements and quality assessments on real
commissioning and test setups, as well as through surveys of end users on user
acceptance.
The following limitations should be noted. Due to shipyard restrictions, con-
tact with the end users has only taken place through the experts at the shipyards
and not directly through the authors. Furthermore, no evaluations have yet taken
place that would allow statistical evaluation through an evaluation concept with
constant boundary conditions and a sufficient number of participants. The fact
that the system has so far only been implemented as a prototype also entails lim-
itations. For a real implementation in the IT infrastructure of the shipyards and
the connection to the ship automation during commissioning, questions regard-
ing security and cybersecurity would have to be clarified. These could not be
fully clarified within the scope of this work.

5 Summary and Outlook

In this paper, a digital assistance system for performing maritime commissioning

is presented. For this purpose, the commissioning process was first recorded with
the support of experts from three shipyards, and current deficits were identified.
62 T. M. Jansen et al.

Based on this and on an existing digital assistance system and digital twin for the
authoring of commissioning tasks, necessary functionalities and user interfaces
were developed and presented. Finally, the resulting potentials were highlighted
and the limitations explained.
If the potential described in the previous chapter can be realized, the concept
will be transferred to industry to achieve long-term integration. As shipbuilding
is an industry that involves several companies and in which suppliers play a
major role [1], the involvement of suppliers in the digital process is crucial for
a successful and holistic digital transformation. The digital assistance system
could be extended to allow, for example, the connection of suppliers and the
inclusion of factory acceptance tests.

Acknowledgement. The authors gratefully acknowledge the financial support with

funds provided by the Federal Ministry for Economic Affairs and Climate Action
(BMWK) due to an enactment of the German Bundestag under Grant No. 03SX503D.

References
1. Strandhagen, J.W., Buer, S.-V., Semini, M. et al.: Sustainability challenges and
how Industry 4.0 technologies can address them: a case study of a shipbuilding
supply chain. Prod. Plann. Control 33, 995–1010 (2022). https://doi.org/10.1080/
09537287.2020.1837940
2. Willner, O., Powell, D., Gerschberger, M., et al.: Exploring the archetypes of
engineer-to-order: an empirical analysis. Int. J. Oper. Prod. Manage. 36, 242–264
(2016). https://doi.org/10.1108/IJOPM-07-2014-0339
3. Birkie, S.E., Trucco, P.: Understanding dynamism and complexity factors in
engineer-to-order and their influence on lean implementation strategy. Prod. Plann.
Control 27, 345–359 (2016). https://doi.org/10.1080/09537287.2015.1127446
4. Mello, M.H., Gosling, J., Naim, M.M., et al.: Improving coordination in an
engineer-to-order supply chain using a soft systems approach. Prod. Plann. Control
28, 89–107 (2017). https://doi.org/10.1080/09537287.2016.1233471
5. California Commissioning Collaborative: California Commissioning Guide: New
Buildings (2006)
6. IPENZ: Practice Note 09: Commissioning Capital Plant (2007)
7. Mills, E.: Building commissioning: a golden opportunity for reducing energy costs
and greenhouse gas emissions in the United States. Energ. Effi. 4, 145–173 (2011).
https://doi.org/10.1007/s12053-011-9116-8
8. Bundesministerium für Wirtschaft und Klimaschutz. https://www.mittelstand-dig
ital.de/MD/Navigation/DE/Themen/Mensch-Digitalisierung/Assistenzsysteme/
assistenzsysteme.html. Accessed 05 June 2023
9. Link, M., Hamann, K.: Einsatz digitaler Assistenzsysteme in der Produktion:
Gestaltung der Mensch-Maschine Interaktion. ZWF Zeitschrift für wirtschaftlichen
Fabrikbetrieb 114, 683–687 (2019). https://doi.org/10.3139/104.112161
10. Hold, P., Erol, S., Reisinger, G., et al.: Planning and evaluation of digital assistance
systems. Procedia Manuf. 9, 143–150 (2017). https://doi.org/10.1016/j.promfg.
2017.04.024
11. Mark, B.G., Rauch, E., Matt, D.T.: Worker assistance systems in manufacturing:
a review of the state of the art and future directions. J. Manuf. Syst. 59, 228–250
(2021). https://doi.org/10.1016/j.jmsy.2021.02.017
 Digital Maritime Commissioning Assistant 63

12. Keller, T., Bayer, C., Bausch, P., et al.: Benefit evaluation of digital assistance
systems for assembly workstations. Procedia CIRP 81, 441–446 (2019). https://
doi.org/10.1016/j.procir.2019.03.076
13. CxPlanner Commissioning Process Software. https://cxplanner.com/
commissioning-process-software. Accessed 17 Apr 2023
14. Whole Building Design Guide. https://www.wbdg.org/building-commissioning.
Accessed 17 Apr 2023
15. Swiss Commissioning. https://swisscommissioning.com/
inbetriebnahmemanagement-software-inbetriebnahme-software/. Accessed 07
Apr 2023
16. Bluerithm. https://bluerithm.com/. Accessed 07 Apr 2023
17. Elzalabany, A., Settler, V., Rost, R., et al.: Digitalization of maritime commission-
ing: a system for automatic content creation and augmented reality-based com-
missioning execution. In: Bertram, V. (ed.) Computer and IT Applications in the
Maritime Industries 2022, COMPIT, pp. 323–335. Hamburg University of Tech-
nology (2022)
18. Giering, J.-E., Dyck, A.: Maritime Digital Twin architecture. at - Automatisierung-
stechnik 69, 1081–1095 (2021). https://doi.org/10.1515/auto-2021-0082
19. Elzalabany, A., Settler, V., Jansen, T., et al.: Automated generation and low-effort
authoring of commissioning content in the maritime industry. In: International
Conference on Computer Applications in Shipbuilding 2022, ICCAS, pp. 85–96.
The Royal Institution of Naval Architects, London (2022)
20. Südekum, N.: Schiffbauoptimierung durch Standardisierung. Schiff&Hafen 72, 12–
13 (2020)
 Transforming Engineer-to-Order
Projects, Supply Chains, and Ecosystems
 Challenges and Opportunities of Software-Based
Production Planning and Control
for Engineer-to-Order Manufacturing

Patrick Bründl1(B) , Micha Stoidner1 , Huong Giang Nguyen1 ,

Andreas Baechler2 , and Jörg Franke1
1 Institute for Factory Automation and Production Systems (FAPS),
Friedrich-Alexander-Universität Erlangen-Nürnberg, Nuremberg, Germany
[email protected]
2 Rittal GmbH & Co. KG, Herborn, Germany

Abstract. This article aims to explore the use and implementation of software-
based production planning and control (PPC) systems in variant-rich manufactur-
ing environments, specifically in the engineer-to-order (ETO) manufacturing. A
systematic literature review is conducted to identify, which PPC tools are used in
the ETO business, how they should be implemented and the benefits and challenges
associated with their use. The results indicate that such systems can be technically
and economically advantageous, however application of existing systems off-the
shelf solutions is not recommended. The literature suggests a structured approach
for the implementation considering critical success factors and a well-planned
process supported by the proposed framework. Most of the systems analyzed in
the literature are Enterprise Resource Planning systems. Furthermore, the findings
show four different areas of positive effects for the companies, such as improved
production efficiency, improved planning efficiency, improved decision-making,
and increased competitiveness. Additionally, five different kinds of challenges
can be observed: Knowledge and expertise, Cost and resource management, Pro-
cess and system flexibility, Data management and IT infrastructure and System
usability and applicability.

Keywords: Engineer-to-Order · Production planning and control · Systematic

literature review · Implementation guidelines

1 Introduction

Manufacturing companies are under increasing pressure to produce high-quality prod-

ucts while meeting the unique needs and demands of their customers. Manufactur-
ing companies operating under engineer-to-order (ETO) strategies are increasingly
faced with the challenge of delivering highly customized products efficiently and cost-
effectively. The dynamic and complex nature of production planning and control (PPC)
in these contexts makes it a nuanced field of study, requiring tailored methodologies.

© IFIP International Federation for Information Processing 2023

Published by Springer Nature Switzerland AG 2023
E. Alfnes et al. (Eds.): APMS 2023, IFIP AICT 691, pp. 67–79, 2023.
https://doi.org/10.1007/978-3-031-43670-3_5
68 P. Bründl et al.

Indeed, strategies successful in other contexts, such as make-to-stock or make-to-order,

may not provide the same benefits when applied to ETO environments [1], emphasizing
the need for dedicated research in this area [2]. This research paper endeavors to address
this academic need by exploring the critical issues and challenges associated with imple-
menting software-based PPC tools in ETO environments and developing a framework to
implement these tools. The specific context of high customization and short lead times,
often leading customers to choose local over overseas manufacturers, will be considered
[2].
A systematic literature review will serve as the research methodology, enabling a
comprehensive exploration of the topic. The primary objective of this study is to enhance
the scientific understanding of production planning in ETO environments. This will
be achieved by investigating which types of software-based PPC tools are commonly
employed, how these systems should be implemented, and what potential benefits and
challenges must be contemplated. As a result, this research aims to provide a significant
contribution to the literature on PPC in complex ETO environments. From a practical
standpoint, the insights gained from this research will also offer valuable guidance for
practitioners. The study provides actionable recommendations for those interested in
implementing software-based PPC systems in ETO settings, contributing to more effec-
tive and efficient production processes. Consequently, while the main emphasis of this
research lies in advancing academic understanding, its findings also have relevant man-
agerial implications, providing a bridge between theory and practice in the realm of ETO
manufacturing.

2 Background

2.1 Engineer-To-Order Production Strategy

ETO is a production strategy particularly suitable for manufacturing highly customized,

non-repetitive products, designed and manufactured in line with unique customer
requirements. This approach is beneficial in environments demanding a high level of
customization, such as control cabinet manufacturing, where products are engineered
from scratch or re-engineered based on existing designs [3]. According to Bertrand,
Muntslag [5] the ETO strategy comprises two primary stages. First, the non-physical
stage encompasses tendering, engineering, design, and process planning activities. Sec-
ond, the physical stage involves component manufacturing, assembly, and installation
[4, 6, 7]. ETO supply chains are dynamic and complex, characterized by the order pene-
tration point located at the design stage [4, 8]. This system is entirely driven by customer
orders, creating a high level of uncertainty in terms of product specification, supply and
delivery lead times, and duration of the production processes [4]. In the contemporary
manufacturing landscape, ETO companies face increasing demands to reduce lead time
and costs while simultaneously providing high levels of customization and flexibility.
Achieving this balance is a significant challenge due to the inherent complexity and
uncertainty of the ETO environment [9, 10]. However, a well-managed PPC process can
enable ETO companies to satisfy customers’ needs while maintaining a high level of
efficiency [11].
 Challenges and Opportunities of Software-Based Production Planning 69

2.2 Software-Based Production Planning and Control

This section presents common PPC systems to provide a theoretical basis for the follow-
ing sections. Enterprise Resource Planning (ERP) systems are comprehensive software
solutions helping companies automate, integrate and optimize core business processes.
Acting as a central data hub, they provide insights for various departments. Advan-
tages include increased productivity, transparency and flexibility. ERP systems comprise
interconnected modules sharing a common database. [12, 13].
Manufacturing Execution Systems (MES) were developed in the 1980s to support
manufacturing activities, bridging planning systems like ERP and control systems. They
utilize manufacturing information to streamline processes. In smart factories, MES offers
real-time information and decision support, including a comprehensive view of pro-
duction resources and support activities like equipment maintenance and production
planning. MES also facilitates real-time data management and production optimization
[14].
Advanced Planning and Scheduling (APS) systems provide a complete view of the
supply chain, enabling improvements through flexible and modular implementation.
They map a company’s entire supply chain setup, facilitating optimization and cost
planning. Therefore, quantity and demand planning within the whole supply chain is
possible including capacity planning and detailed scheduling. APS systems calculate
alternative supply chain process options and their costs, links and optimizes the work
of people, and automates processes. Since they are widely used in global supply chains,
APS systems often use cloud technology to be accessible from anywhere [15, 16].
Workload Control (WLC) is a PPC concept that enables simultaneous control of
order input and capacity output, aiming to manage manufacturing bottlenecks and reduce
work-in-progress. It uses a pre-shop pool and order release mechanism, holding orders
to avoid bottlenecks and accommodate unexpected changes. Despite gaining attention
since the 1980s, WLC has seen limited implementation due to the gap between theory and
practice and resistance to change resulting from a lack of understanding and awareness
[17].

3 Research Aim
ETO manufacturing environments present distinct challenges that demand specialized
research attention. While manufacturing companies are under increasing pressure to
deliver high-quality, cost-effective products, the dynamic nature of PPC in ETO envi-
ronments necessitates unique methodologies. Strategies that work well in make-to-stock
or make-to-order environments may not yield the same benefits when applied to ETO
contexts [1]. This underscores the scientific need for dedicated research in this area [2].
ETO producers, in their quest for achieving a unique selling proposition, need to focus
on the internal process optimization, such as digitization and optimization of business
processes, to enhance their service level to end-customers [18].
Improvement in the productivity of ETO production can be facilitated by restruc-
turing both the production process and production management [19]. Guided by these
insights, this research paper endeavors to provide a systematic overview of the benefits
and challenges related to software-based PPC implementation in ETO environments. It
70 P. Bründl et al.

also aims to establish a framework for successfully deploying ETO-tailored systems. To

achieve this, the following research questions will be explored:
• RQ1: What types of software-based PPC tools are commonly used in the ETO
environment?
• RQ2: How can PPC systems be implemented in the production of complex products?
• RQ3: What are the potential benefits of implementing software-based PPC systems
and how can these benefits be evaluated?
• RQ4: What are the technological and economic challenges associated with imple-
menting PPC systems in an ETO environment?
The answers to these questions, obtained through a systematic literature review, will
form a framework for implementing ETO-appropriate systems and provide a comprehen-
sive classification of implementation factors. Additionally, this study aims to compile
a list of economic and technological benefits and challenges associated with imple-
menting these systems, addressing a significant gap in the current literature. While the
primary focus of this research is to advance academic understanding, the findings will
also bear significant managerial implications. The study will offer valuable guidance
for practitioners seeking to implement software-based PPC systems in ETO settings,
thereby contributing to more effective and efficient production processes. This research
is expected to highlight further areas of research need, ultimately contributing to the
continuous development of this complex and evolving field.

4 Methodology
The methodology chosen for this study is a systematic literature review, which provides
the opportunity to evaluate and synthesize the existing literature with minimal bias, high
efficiency and consistency. The study follows the methodology of Tranfield et al. [20]
to ensure transparency, completeness and reproducibility in the review process. The
selection of literature will be limited by several factors to ensure clear review paths
[21–23].
Timeline: The study considers literature published between January 2000 and July
2022 for the literature review. The year 2000 has been selected as the starting point,
marking the definition of the initial part of the ISA-95 standard, which was the first
to define models and transactions for integrating software-based systems for PPC [24].
These integrated systems form the basis of modern PPC systems, which are the primary
subject of investigation in this study. The endpoint of July 2022 has been selected to
ensure the stability of the sample during the analysis, preventing any alterations to the
dataset within this period.
Databases: The study identified Scopus, Business Source Complete (EBSCO) and Web
of Science as relevant databases for the literature review.
Publication Types: Only peer-reviewed journal articles and conference proceedings
were included in the study.
Selection of Articles: The resulting string used for queries in the mentioned databases
can be divided into two blocks.
 Challenges and Opportunities of Software-Based Production Planning 71

Block 1: (“Production Management System*” OR “Production Control Sys-

tem*” OR “Planning and scheduling” OR “Production planning” OR PPC)
Block 2: (varied OR “many variation*” OR variant* OR “rich variety” OR “full of
variety” OR variable* OR versatile* OR “wide ranging” OR “small number*” OR
“low number*” OR “small order*” OR “small volume*” OR “small production
lot*” OR “small batch size*” OR “small lot*” OR “low quantities” OR “lot size*
1” OR “lot-size 1” OR “batch size 1” OR “batch-size 1” OR “size of the lot*”
OR “lot size one” OR “lot-size one” OR “batch size one” OR “batch-size one”
OR “engineer-to-order” OR “ETO” OR “customized” OR “custom-made” OR
“custom-built” OR “one-of-a-kind-production” OR OKP)
The search based on the queries and the selected parameters resulted in 3599 hits.
After screening the titles and removing duplicates, the number of relevant literature
was reduced to 334. After removing duplicates and examining the abstracts, 124
articles were considered results. During this process, it was ensured that the articles
discussed software-based production planning and control and pertained to variant-
rich production environments. In the end, 31 articles were used as the basis for
this study.

Classification: The authors independently reviewed and categorized all 31 results in

detail, extracting relevant subtopics. Following this, the categorizations were discussed,
and a consensus on the content was reached, following the guidelines and measures
suggested by Durach et al. [25] to minimize the within-study bias and the expectancy
bias.
Evaluation: The categorized articles were discussed and examined for consistency and
potential contradictions. The implications were critically evaluated to determine if they
were adequately supported or if further research was needed to confirm the findings.
This review allowed for the research question posed in the first chapter to be answered.

5 Findings and Discussion

5.1 Types of PPC Tools in ETO Environments

Figure 1, illustrating the results of the literature analysis, shows that ERP systems with
Material Requirements Planning (MRP II) algorithms are the most commonly described
systems for PPC in ETO companies. Despite their prevalence, the literature often points
out that these systems are not entirely adequate for successful PPC in ETO settings.
Following ERP systems, the second most frequently mentioned tools are Excel and
Visual Basic for Applications (VBA) macros for planning. The literature also explores
various proprietary solutions, including multiagent systems and dedicated platforms.
APS systems are notable for considering both upstream and downstream value chain
stages.
Interestingly, most systems, including proprietary ones, tend towards a pull orien-
tation, even though only one explicit mention of a classical Kanban system was found.
MES are rarely mentioned in the literature, with only a single reference discussing
72 P. Bründl et al.

20
18
Number of contributions 16
14
12
10
8
6
4
2
0

Fig. 1. Overview of the presented PPC systems

a PPC system under the MES term. Another system, based on the constant work in
process (CONWIP) approach, also represents a pull-based system, providing more flex-
ibility than a traditional Kanban system [26]. The wide array of potential computer-based
systems for PPC optimization in variant-rich manufacturing is reflected in the literature.
It’s important to note that articles dealing with complex systems like APS and MES
always address the integration of these solutions with the ERP system as a backbone.
However, the frequency of mentions of APS and MES systems does mirror the maturity
level or actual implementation of these systems in ETO industries. This correlates with
a survey of Czech manufacturing companies revealing that 48% use ERP systems with
MRP II algorithms, followed by 23% using Excel or similar proprietary solutions [27].
Additional literature like Hendry et al. [28], also emphasize the vast majority of ERP
systems and a lack of sophisticated systems for application in ETO environments. This
can therefore still be confirmed.

5.2 Implementation of PPC Tools

The sum of the functions and stakeholders involved in the various systems places high
demands on an implementation project. Therefore, a detailed requirements analysis
should be performed at the beginning for the functions required [3, 29, 30]. The most
frequently described functions in the literature include supporting the execution of pro-
duction processes, monitoring production performance and collecting production data.
In addition, order tracking, capacity planning and machine monitoring are notable func-
tions of the systems [3, 31]. The identification of Critical Success Factors (CSFs) can
significantly aid the implementation of a PPC system, as suggested in the literature [32].
To effectively display the factors described in the literature, a framework that categorizes
these factors into the three dimensions: human, technological, and organizational is used.
This categorization, used by Petter et al. [33] and Invernizzi et al. [34], provides a com-
prehensive and systematic approach to understanding and managing the complexities
involved in the implementation of a PPC system.
Human Factors: Critical elements for a successful implementation include the compo-
sition of the project team, robust communication, thorough education and training, and
 Challenges and Opportunities of Software-Based Production Planning 73

active user involvement. Given that companies often lack the necessary IT and project
management expertise internally [35], it is recommended to bring this knowledge into
the project team through external consultants [36]. Seamless communication among
company management, project and technical teams, employees, and other stakehold-
ers is considered a cornerstone for successful implementation [29, 37, 38]. Ensuring
comprehensive staff education and training is also crucial [28, 36, 37]. User partici-
pation in the system development and implementation process is equally significant,
as it improves the quality of the enterprise system, making it more user-friendly [37].
This becomes particularly essential as many planning steps were previously performed
manually, meaning that individual employees possess a wealth of implicit knowledge.
Through a collaborative approach, this untapped knowledge can be utilized, benefitting
the company [19, 28, 39–41].
Technological Factors: During the software selection process, it is essential to pay
close attention to defining the requirements that the chosen PPC system must satisfy,
while also ensuring compatibility with existing enterprise systems. The technological
infrastructure holds central importance as the implementation signifies a major alteration
in the current information systems and business processes at both physical and control
levels [42]. An advanced information and communication technology infrastructure is
required to guarantee a smooth flow of information [43]. To ensure the performance of
the system, attention should be paid to the selection of the required data and the degree
of accuracy of these, especially at the beginning when adequate data might not be readily
available [28]. Many tasks are often managed using standalone solutions or legacy tools,
leading to isolated workflows [3, 44]. Therefore, in designing the architecture, it is vital to
provide the necessary interfaces and to achieve comprehensive integration of all involved
processes and systems [30, 36, 45].
Organizational Factors: The involvement of top management is crucial in several
aspects, such as resource allocation, partner selection, management of user resistance,
and encouragement of participation [29, 31, 37, 40, 46]. Solid project management
encompasses consistent management of the rollout plan, defining tasks, and establishing
interdepartmental collaboration. Besides managing the project, top management must
also display a strong commitment to the rollout project and the necessary change pro-
cesses [38]. Change management strategies are aimed at administering the widespread
cultural and structural modifications within the company. These are essential to engage
employees in the project and facilitate the transition [28, 38].
With the help of the literature, the following approach can be suggested from the
points described. The necessary process changes can be designed through business pro-
cess reengineering to optimize the existing processes [36]. Finally, the measurement and
evaluation of the implementation are also essential as monitoring the progress contributes
to the acceptance and improvement of the system [34]. The overall implementation
framework can be summarized into four main work packages: implementation prepa-
ration, business process redesign, system development, as well as testing and training.
Before implementing a system, three preparation tasks should be completed: Assem-
ble a project team, establish a clear project plan with specific milestones, describe data
74 P. Bründl et al.

requirements, collect data, and design data capture capabilities for the processes and
interfaces from related systems.
To take full advantage of the system’s capabilities, the original process flow needs
to be redesigned with the help of the involved stakeholders. Once the system imple-
mentation is complete, an extensive testing phase follows. Based on the test reports,
improvements are made to the system and manuals, guidelines and training materials
are created. Training is then conducted in a live environment. This feedback is summa-
rized and discussed by the project team and further improvements are made based on
the feedback [37].

5.3 Potential Benefits of PPC Tools

Most of the described optimization possibilities are described qualitatively and not sup-
ported quantitatively with respect to their positive effects, as further research and dissem-
ination of the systems are needed. Nevertheless, the research results show first benefits of
system implementation. These benefits can be categorized into the following categories.
Software PPC systems offer comprehensive benefits for ETO companies, improv-
ing operational and planning efficiency, enhancing competitiveness, and supporting
decision-making processes. The positive impact on production efficiency is a recur-
ring theme in the literature. These systems boost productivity [39, 41, 42, 46, 47] and
enhance flexibility and responsiveness [3, 19, 36, 40, 42, 43, 45]. Importantly, automa-
tion becomes more feasible [42, 51], additionally catalyzing efficiency in processes and
lead times [36, 48–51]. Planning efficiency also experiences significant improvement.
These systems facilitate more precise capacity and requirements planning [39, 52] and
foster collaborative planning across different departments and systems [19, 31, 36, 53].
Automation of planning tasks further streamlines the planning process [44, 49, 51, 53,
54]. A crucial aspect is the realistic setting of due dates [19, 28, 29, 38, 47–50] and
reducing errors, rework, and downtime [19, 38, 42, 47].
The competitiveness of ETO companies is additionally bolstered through lower man-
ufacturing costs [39, 42, 43, 52, 55] and reduced inventories and work in progress [3, 40,
48–50]. This results in increased customer satisfaction [29, 47, 50, 55] and notably, an
improved quotation process [28]. The systems also support improved decision-making
by enhancing transparency and process monitoring [19, 31, 38, 40, 51, 53, 55]. They
equip companies with data-driven insights [31, 42, 55] that feed into more informed
decision-making. A vital benefit of these systems is their capacity to improve the quota-
tion process. In ETO industries, which often operate within a tendering process, a PPC
tool’s ability to accurately assess order and capacity situations becomes invaluable. It
enables the provision of more realistic quotes and due dates. This, in turn, enhances
customer satisfaction by mitigating the risk of delivery delays or unexpected cost esca-
lations. Furthermore, it promotes smoother production levelling within the company,
contributing to operational efficiency.
This benefit also resonates with the order-to-delivery aspect, which is one of the
most frequently cited themes in the literature. It underscores the pivotal role of software
production planning and control systems in streamlining the order-to-delivery cycle,
thereby heightening the overall operations and competitiveness of ETO companies.
 Challenges and Opportunities of Software-Based Production Planning 75

5.4 Technological and Economic Challenges

Based on the literature reviewed, the challenges of using PPC systems in the ETO
environment can be classified into five categories, which are discussed below.
Implementing PPC systems in ETO environments poses a set of distinct challenges.
The “Knowledge and expertise” sector is pivotal, as several studies highlight the shortage
of necessary skills during the selection and implementation stages [3, 31, 35, 38, 41,
49], as well as the need for further research and real-life application [3, 28, 36–39].
The “Cost and resource management” hurdle stems from the cost-intensive nature of
PPC implementation and change management [31, 38, 41, 44, 56], further amplified
by the high expenses linked to acquisition, customization [31, 35, 41, 44, 50, 53, 55].
The inherent characteristic of ETO environments to deal with unique, custom products
necessitates high “Process and system flexibility”.
This demand is intensified by uncertainties and variations in planning [19, 29, 30,
36, 38, 39, 41–45, 47, 48, 53–55] and unformalized and unstructured processes [3, 29,
30, 35, 37, 41, 53], the need for cross-departmental coordination and integration [3, 19,
24, 30, 31, 42, 43, 45, 48, 51, 53, 56], and product and process variations [41, 45, 47, 54,
56]. Data-related challenges reside within the “Data management and IT infrastructure”
realm. Obstacles such as missing information at the enquiry stage [28, 30, 31, 37],
the presence of legacy systems and data silos [3, 19, 24, 28, 31, 36, 44, 50, 51, 53,
56], and the lack of homogeneous data structures prevent the creation of an up-to-date,
integrated data infrastructure [28, 36, 39, 41, 43, 47, 49, 51, 54, 55], while impeding the
implementation of effective data capturing methods [36, 40, 43, 51].
Lastly, “System usability and applicability” issues arise due to the complexity of
PPC systems in ETO contexts and the limited applicability of current PPC systems
for ETO [2, 3, 19, 29–31, 35, 36, 41, 46, 47, 54]. An overarching challenge in ETO
environments includes dealing with uncertainty brought on by manual processes and data
deficiency, accentuated in planning and scheduling where lack of accurate data causes
traditional solutions to fall short. Companies in the ETO sector need to strategize and
invest in appropriate resources to implement PPC systems successfully. This process calls
for careful system selection and customization, strong change management, employee
training, continuous research and real-world application to refine algorithms, improve
system usability, and develop adaptable, integrated processes. By doing so, they can
mitigate many of the challenges associated with ETO production planning and control.

6 Conclusions

The literature analysis shows that implementing software-based PPC systems in ETO
environments can provide technical and economic benefits. However, it is important to
avoid the application of pre-existing systems without considering the specific charac-
teristics of ETO environments. Instead, a systematic approach is recommended, which
encompasses organizational, human, and technological factors. A particular emphasis
is placed on comprehensive planning and preparation during the implementation stage,
where the presented implementation framework can be utilized for the iterative improve-
ment of the software-driven PPC process. Key factors for success include putting together
76 P. Bründl et al.

a proficient project team, making a thoughtful selection of the software, and obtaining
broad organizational support.
The systems most commonly analyzed in the literature are ERP systems, closely
followed by Excel-based solutions and various in-house alternatives. Furthermore, APS
is becoming increasingly popular in the ETO sector. Even with the limited availability
of quantitative data on the benefits of these systems, the analysis suggests that posi-
tive effects can be experienced across four main areas. This finding carries substantial
managerial implications. Despite the challenges in assessing PPC systems for high-
variability manufacturing, it is crucial for companies to stay abreast of current develop-
ments to maintain their competitive edge. Therefore, industries working in ETO should
consider the introduction of a PPC system, considering their unique starting situation,
the suitability of the system, and the guidelines for implementation.
This analysis also implies significant theoretical implications, underlining the need
for further research. Such research should aim to identify technological solutions to the
issues presented by PPC systems, as well as to validate the benefits of these systems
using real-world industry data. One specific area of research that needs attention is the
development of a framework for integrated models that address all aspects of production
planning and control in ETO environments. This future research must be firmly grounded
on industrial data to ensure its relevance and applicability. Additionally, it is critical to
integrate new technological approaches into this research, with the aim of creating an
integrated system that puts the product and its specifications at the focus. By doing so, we
can ensure the development of a system that is not only practical and effective, but also
specifically tailored to the unique requirements and circumstances of ETO industries.

References
1. Rahman Abdul Rahim, A., Shariff Nabi Baksh, M.: The need for a new product development
framework for engineer-to-order products. Eur. J. Innov. Manag. 6, 182–196 (2003). https://
doi.org/10.1108/14601060310486253
2. Sriram, P.K., Alfnes, E.: Taxonomy of engineer-to-order companies. In: Grabot, B., Valle-
spir, B., Gomes„ S., Bouras, A., et al. (eds) Advances in Production Management Systems.
Innovative and Knowledge-Based Production Management in a Global-Local World. APMS
2014. IFIP Advances in Information and Communication Technology, vol. 8827, pp. 579–587.
Springer, Heidelberg (2014)
3. Adrodegari, F., Bacchetti, A., Pinto, R., et al.: Engineer-to-order (ETO) production planning
and control: an empirical framework for machinery-building companies. Prod. Plann. Control
26, 910–932 (2015). https://doi.org/10.1080/09537287.2014.1001808
4. Wikner, J., Rudberg, M.: Integrating production and engineering perspectives on the customer
order decoupling point. Int. J. Op Prod. Manage. 25, 623–641 (2005). https://doi.org/10.1108/
01443570510605072
5. Bertrand, J., Muntslag, D.R.: Production control in engineer-to-order firms. Int. J. Prod. Econ.
30–31, 3–22 (1993). https://doi.org/10.1016/0925-5273(93)90077-X
6. Amaro, G., Hendry, L., Kingsman, B.: Competitive advantage, customisation and a new
taxonomy for non make-to-stock companies. Int. J. Op Prod. Manage. 19, 349–371 (1999).
https://doi.org/10.1108/01443579910254213
7. Gosling, J., Naim, M.M.: Engineer-to-order supply chain management: a literature review
and research agenda. Int. J. Prod. Econ. 122, 741–754 (2009). https://doi.org/10.1016/j.ijpe.
2009.07.002
 Challenges and Opportunities of Software-Based Production Planning 77

8. Gosling, J., Hewlett, B., Naim, M.M.: Extending customer order penetration concepts to
engineering designs. Int. J. Op. Prod. Manage. 37, 402–422 (2017). https://doi.org/10.1108/
IJOPM-07-2015-0453
9. Gosling, J., Towill, D.R., Naim, M.M., et al.: Principles for the design and operation of
engineer-to-order supply chains in the construction sector. Prod. Plann. Control, 1–16 (2014).
https://doi.org/10.1080/09537287.2014.880816
10. Birkie, S.E., Trucco, P.: Understanding dynamism and complexity factors in engineer-to-order
and their influence on lean implementation strategy. Prod. Plann. Control 27, 345–359 (2016).
https://doi.org/10.1080/09537287.2015.1127446
11. Chase, R., Jacobs, R., Aquilano, N.: Operations Management for Competitive Advantage (mit
CD), 10th edn. McGraw Hill, s.l. (2004)
12. Matolcsy, Z.P., Booth, P., Wieder, B.: Economic benefits of enterprise resource planning
systems: some empirical evidence. Account. Financ. 45, 439–456 (2005). https://doi.org/10.
1111/j.1467-629X.2005.00149.x
13. Shang, S., Seddon, P.B.: A comprehensive framework for classifying the benefits of ERP
systems. In: Association for Information Systems (ed.) AMCIS 2000 Proceedings, 39th edn.,
pp. 1005–1014 (2000)
14. Saenz de Ugarte, B., Artiba, A., Pellerin, R.: Manufacturing execution system – a litera-
ture review. Prod. Plann. Control 20, 525–539 (2009). https://doi.org/10.1080/095372809029
38613
15. Steger-Jensen, K., Hvolby, H.-H., Nielsen, P., et al.: Advanced planning and scheduling tech-
nology. Prod. Plann. Control 22, 800–808 (2011). https://doi.org/10.1080/09537287.2010.
543563
16. Mauergauz, Y.: Advanced Planning and Scheduling in Manufacturing and Supply Chains.
Springer International Publishing, Cham, s.l (2016)
17. Thürer, M., Stevenson, M., Silva, C.: Three decades of workload control research: a systematic
review of the literature. Int. J. Prod. Res. 49, 6905–6935 (2011). https://doi.org/10.1080/002
07543.2010.519000
18. Pfeifer, M.R.: Development of a smart manufacturing execution system architecture for smes:
a czech case study. Sustainability (Switzerland) 13 (2021). https://doi.org/10.3390/su1318
10181
19. Nujen, B.B., Alfnes, E., Mwesiumo, D., et al.: Centralized vs decentralized production plan-
ning in ETO environments: a theoretical discussion. In: Kim, D.Y., Cieminski G von, Romero,
D. (eds.) Advances in Production Management Systems. Smart Manufacturing and Logistics
Systems: Turning Ideas into Action, vol 664, pp 329–338. Springer, Cham (2022)
20. Tranfield, D., Denyer, D., Smart, P.: Towards a methodology for developing evidence-
informed management knowledge by means of systematic review. Br. J. Manage. 14, 207–222
(2003). https://doi.org/10.1111/1467-8551.00375
21. Rojon, C., Okupe, A., McDowall, A.: Utilization and development of systematic reviews in
management research: What do we know and where do we go from here? Int. J. Manag. Rev.
23, 191–223 (2021). https://doi.org/10.1111/ijmr.12245
22. Mulrow, C.D.: Rationale for systematic reviews. BMJ 309, 597–599 (1994). https://doi.org/
10.1136/bmj.309.6954.597
23. Soni, G., Kodali, R.: A critical analysis of supply chain management content in empirical
research. Bus. Process. Manag. J. 17, 238–266 (2011). https://doi.org/10.1108/146371511
11122338
24. Nigischer, C., Hennig, M., Gerhard, D.: A concept to integrate manufacturing execution and
product data management systems. In: Fortin, C., Rivest, L., Bernard, A., et al. (eds.) Product
Lifecycle Management in the Digital Twin Era, vol. 565, pp. 315–324. Springer, Cham (2019)
25. Durach, C.F., Kembro, J., Wieland, A.: A new paradigm for systematic literature reviews in
supply chain management. J Supply Chain Manag 53, 67–85 (2017).
78 P. Bründl et al.

26. Framinan, J.M., González, P.L., Ruiz-Usano, R.: The CONWIP production control system:
review and research issues. Prod. Plann. Control 14, 255–265 (2003). https://doi.org/10.1080/
0953728031000102595
27. Hrušecká, D.: Process innovation as a necessary condition for successful implementation of a
new production planning system. In: Proceedings of the European Conference on Innovation &
Entrepreneurship, pp. 514–521 (2014)
28. Hendry, L., Land, M., Stevenson, M., et al.: Investigating implementation issues for workload
control (WLC): a comparative case study analysis. Int. J. Prod. Econ. 112, 452–469 (2008).
https://doi.org/10.1016/j.ijpe.2007.05.012
29. Cannas, V.G., Pero, M., Pozzi, R., et al.: An empirical application of lean management tech-
niques to support ETO design and production planning. IFAC-PapersOnLine 51, 134–139
(2018). https://doi.org/10.1016/j.ifacol.2018.08.247
30. Little, D., Rollins, R., Peck, M., et al.: Integrated planning and scheduling in the engineer-to-
order sector. Int. J. Comput. Integr. Manuf. 13, 545–554 (2000). https://doi.org/10.1080/095
11920050195977
31. Persona, A., Regattieri, A., Romano, P.: An integrated reference model for production planning
and control in SMEs. J. Manuf. Technol. Manag. 15, 626–640 (2004). https://doi.org/10.1108/
17410380410555871
32. Sjøbakk, B., Bakås, O., Bondarenko, O., et al.: Designing a performance measurement system
to support materials management in engineer-to-order: a case study. Adv. Manuf. 3, 111–122
(2015). https://doi.org/10.1007/s40436-015-0109-2
33. Petter, S., DeLone, W., McLean, E.R.: Information Systems Success: The Quest for the
Independent Variables. J. Manag. Inf. Syst. 29, 7–62 (2013). https://doi.org/10.2753/MIS
0742-1222290401
34. Invernizzi, D., Gaiardelli, P., Arica, E., et al.: MES implementation: critical success factors
and organizational readiness model. IFIP Adv. Inf. Commun. Technol. 567, 493–501 (2019).
https://doi.org/10.1007/978-3-030-29996-5_57
35. Adrodegari, F., Bacchetti, A., Sicco, A., et al.: One-of-a-Kind Production (OKP) Planning and
Control: An Empirical Framework for the Special Purpose Machines Industry. In: Emmanoui-
lidis, C., Taisch, M., Kiritsis, D. (eds.) Advances in Production Management Systems. Com-
petitive Manufacturing for Innovative Products and Services, vol. 398, pp. 630–637. Springer,
Heidelberg (2013)
36. Gore, A., Haapasalo, H., Kess, P.: Production planning through customized ERP at a nordic
manufacturing company. Management 6, 211–223 (2011)
37. Huang, Y.: A state-of-the-Art workload control system for customized industry. In: IEEE
International Conference on Industrial Engineering and Engineering Management (2014).
Doi:https://doi.org/10.1109/IEEM.2013.6962602
38. Piller, C., Wölfel, D.W.: Production planning for SMEs – implementation of production plan-
ning with subject-oriented business process management (S-BPM). In: Junqueira Barbosa,
S.D., Chen, P., Cuzzocrea, A., et al. (eds.) S-BPM ONE - Application Studies and Work in
Progress, vol. 422, pp. 164–173. Springer, Cham (2014)
39. Ghiyasinasab, M., Lehoux, N., Ménard, S., et al.: Production planning and project scheduling
for engineer-to-order systems- case study for engineered wood production. Int. J. Prod. Res.
59, 1068–1087 (2021). https://doi.org/10.1080/00207543.2020.1717009
40. Klega, R.: Improvement of production planning based on the application of principles of
MRP. Acta Logistica, pp. 1–7 (2014)
41. Nakayama, R.S., et al.: Production planning and control in small engineer-to-order companies:
understanding difficulties and pragmatic approach. In: Portland International Conference
on Management of Engineering and Technology 2015-September (2015). https://doi.org/10.
1109/PICMET.2015.7273118
 Challenges and Opportunities of Software-Based Production Planning 79

42. Buchmeister, B., Polajnar, A., Palcic, I., et al.: Adaptive control systems for responsive fac-
tories. In: Annals of DAAAM and Proceedings of the International DAAAM Symposium
(2010)
43. Atug, J., Hees, A., Wagner, M., et al.: Production planning for customer innovated products.
In: IEEE International Conference on Industrial Engineering and Engineering Management
2016-December (2016). https://doi.org/10.1109/IEEM.2016.7798014
44. Nam, S., Shen, H., Ryu, C., et al.: SCP-Matrix based shipyard APS design: application to
long-term production plan. Int. J. Naval Architecture Ocean Eng. 10, 741–761 (2018). https://
doi.org/10.1016/j.ijnaoe.2017.10.003
45. Teich, T., Richter, M., Militzer, J., et al.: Integration of process-and production-planning for
SME. In: International Institute of Informatics and Systemics (ed) Proceedings of The 5th
International Symposium on Management, Engineering and Informatics: MEI 2009 (2009)
46. Yang, L.-R.: Key practices, manufacturing capability and attainment of manufacturing goals:
the perspective of project/engineer-to-order manufacturing. Int. J. Project Manage. 31, 109–
125 (2013). https://doi.org/10.1016/j.ijproman.2012.03.005
47. Lenort, R., Klepek, R., Samolejova, A., et al.: Production paths - An Innovative concept for
heavy machinery production planning and control. Metalurgija 53, 78–80 (2014)
48. Gejo García, J., Gallego-García, S., García-García, M.: Development of a pull production
control method for ETO companies and simulation for the metallurgical industry. Appl. Sci.
10, 274 (2020). https://doi.org/10.3390/app10010274
49. Hvolby, H.-H., Steger-Jensen, K.: Technical and industrial issues of Advanced Planning and
Scheduling (APS) systems. Comput. Ind. 61, 845–851 (2010). https://doi.org/10.1016/j.com
pind.2010.07.009
50. Stopper, M., Gastermann, B., Luftensteiner, F., et al.: ConWIP-based packing order planning
software prototype for variant-centered manufacturing. Lecture Notes in Engineering and
Computer Science 2210 (2014)
51. Zhongyi, M., Younus, M., Yongjin, L.: Automated planning and scheduling system for the
composite component manufacturing workshop. Engineering Letters 19 (2011)
52. Toivonen, V., Ikkala, K., Niemi, E., et al.: Planning and scheduling system for make-to-order
production. CIRP Ann. 55, 493–496 (2006). https://doi.org/10.1016/S0007-8506(07)60466-1
53. Zennaro, I., Finco, S., Battini, D., et al.: Big size highly customised product manufacturing
systems: a literature review and future research agenda. Int. J. Prod. Res. 57, 5362–5385
(2019). https://doi.org/10.1080/00207543.2019.1582819
54. Sudhoff, M., Schüler, P., Herzog, M., et al.: Proving the applicability of assembly complexity
measures for process time prediction of customer-specific production. Procedia CIRP 107,
381–386 (2022). https://doi.org/10.1016/j.procir.2022.04.062
55. Alfnes, E., Hvolby, H.-H., Ameri, F., Stecke, K.E., von Cieminski, G., Kiritsis, D.: APS
feasibility in an engineer to order environment. IFIP Adv. Inf. Commun. Technol. 566, 604–
611 (2019). https://doi.org/10.1007/978-3-030-30000-5_74
56. Du, J., Jiao, Y.-Y., Jiao, J.: Integrated BOM and routing generator for variety synchronization
in assembly-to-order production. J. Manuf. Technol. Manag. 16, 233–243 (2005). https://doi.
org/10.1108/17410380510576859
 Towards the Digital Factory Twin
in Engineer-to-Order Industries: A Focus
on Control Cabinet Manufacturing

Micha Stoidner1(B) , Patrick Bründl1 , Huong Giang Nguyen1 ,

Andreas Baechler2 , and Jörg Franke1
1 Institute for Factory Automation and Production Systems (FAPS),
Friedrich-Alexander-Universität Erlangen-Nürnberg, Nuremberg, Germany
[email protected]
2 Rittal GmbH & Co. KG, Herborn, Germany

Abstract. Digital transformation poses a challenge for small and medium-sized

enterprises (SMEs) with an Engineer-to-Order (ETO) production characterized
by a low-variant and low-volume manufacturing philosophy. This article aims to
examine the role of a digital factory twin and material flow simulation to increase
value creation in ETO production. The study follows a structured approach includ-
ing a literature review, on-site process analysis, and simulations to determine the
optimal production process design. The focus of qualitative and quantitative find-
ings is on the control cabinet manufacturing as an important sector in Germany
representative for ETO production. The findings comprise a novel definition of the
digital factory twin (DFT), implementation of a simulation model and integration
of lean manufacturing principles to optimize ETO production of control cabinets.
This study makes a valuable contribution to the development of the DFT and illus-
trates the successful implementation of simulation as well as lean manufacturing
principles.

Keywords: Control Cabinet · Digital Factory Twin · Engineer-to-Order ·

Material Flow Simulation

1 Introduction
The control cabinet manufacturing industry is facing significant challenges in adapting
to the disruptive changes brought about by Industry 4.0 [1]. While this industry is an
important enabler for automation in various sectors, such as machinery, automotive,
aerospace, and electrical engineering, the manufacturing process for control cabinets
still relies heavily on manual labor due to its high level of customization. The process
involves a series of time-consuming and manual tasks, particularly in the assembly and
wiring of various components and equipment [2]. Additionally, the industry is facing
the challenges that every Engineer-to-Order (ETO) production is challenged by, like
changing customer requirements, volatile orders, and lack of scalable processes [3, 4].
© IFIP International Federation for Information Processing 2023
Published by Springer Nature Switzerland AG 2023
E. Alfnes et al. (Eds.): APMS 2023, IFIP AICT 691, pp. 80–95, 2023.
https://doi.org/10.1007/978-3-031-43670-3_6
 Towards the Digital Factory Twin in Engineer-to-Order Industries 81

Companies producing control cabinets are mostly small and medium enterprises (SMEs).
The industry has been slow to adopt digital technologies due to a lack of expertise and
resources [1, 2, 5]. For example, many companies still rely on analogue documents or
handwritten notes in their production and assembly processes. Electricians often must
wade through hundreds of pages of circuit diagrams to find the information they need
to wire, wasting time and making mistakes. To address these challenges, tools such as
material flow simulations can be used to optimize the manufacturing process for control
cabinets. While these simulations have been widely used in various industries, their
application in the control cabinet manufacturing industry is still unexplored. By using
simulations to analyze and optimize the material flow process, companies can reduce
manual labor, improve efficiency and save time and costs [6].
The study follows a structured approach that includes a literature review, on-site pro-
cess analysis and simulations utilizing the “intuitive method1 ” to determine the optimal
production process design [6]. By working out possible applications as well as current
characteristics of a digital factory twin (DFT), this paper allows academia to evaluate
implemented DFT. Ultimately, this research will contribute to the advancement of the
control cabinet manufacturing industry by promoting the adoption of digital technologies
such as the DFT.

2 Background

2.1 Control and Switchgear Cabinet Manufacturing

Control cabinets are used in all relevant sectors of the manufacturing industry, such as
mechanical engineering, automotive, aerospace and process industries. The high number
of variants down to lot size one in control cabinet manufacturing poses a challenge to
the exploitation of economies of scale [2]. In addition, IT integration is hindered by a
lack of capacity and expertise to optimize processes along the value chain [1]. In control
cabinet manufacturing, the value chain can be described by the phases order processing,
development, procurement, manufacturing and after-sales service [2]. A more detailed
overview of the specific process steps in manufacturing is provided by Joppen, Enzberg,
Kühn, Dumitrescu [5], who describe production with the sub-items “work preparation,
mechanical processing, assembly, wiring, assembly, testing and delivery”. Each of these
steps is a complex process that requires many material processing and assembly tasks
[4, 7].
As every customer project must be developed according to a technical specification
based on the customer’s order, the control cabinet industry is a typical example of a
sector that produces engineer-to-order (ETO) [8]. Companies operating ETO are fac-
ing specific challenges like dynamic changes due to incoming orders, but also change
requests formulated by customers after the start of production [7, 9, 10]. Other uncer-
tainties include the quality of the data provided at the time the order is placed and the
estimation of the required capacity and process times for batch size one production [9,
1 The intuitive method is defined by Linda Riley [6] and consists of the steps 1) changing param-
eter levels; 2) completing valid simulation runs and 3) varying the input data and analyze the
results.
82 M. Stoidner et al.

10]. Additionally, many variants and simultaneously increasing demands on costs, time,
and quality are challenges for the companies [2]. Often there is a workshop-oriented
production, a low level of automation combined with a lack of integration of IT systems,
and only little data [1]. To economically manufacture customizable products in batch
size one in a country with high labor costs, it is, therefore, necessary to develop adapted
production solutions that meet the changing and increasingly complex requirements [1,
7].
Control cabinet manufacturers require a comprehensive and holistic view of the
entire value chain to produce highly customized products economically and efficiently
according to ETO. Current practices for ETO production are not sustainable for a long-
term competitive advantage, as exemplified by control cabinet manufacturing. In the
context of digitalization, the concept of a digital factory has shown high potential for
optimization. The concept, development, and implementation of the digital factory for
ETO are barely researched and necessitate qualitative as well as quantitative research.

2.2 Digital Factory

The digital twin (DT) has received increasing research interest in recent years and
the number of publications in the field has grown rapidly [11]. While various mod-
els and frameworks have been introduced and studied, there are few agreed definitions.
Nevertheless, numerous reviews deal with the topic [12–17].
One application of the DT is the Digital Factory, which is described in VDI Guideline
4499 - Part 1 [18]. By integrating consistent data management and data exchange, the
digital factory is one of the first steps towards Industry 4.0. The digital factory focuses
on linking product data from development with planning data at an early stage and on
coordinated production planning and design. It also provides tools for the modelling,
simulation, planning and visualization of production processes. The aim is to compre-
hensively plan, evaluate and improve all structures, processes, and resources of the real
factory, including the product. The planning phase should be accelerated and overlap
with product development. Investment decisions should be digitally validated in advance.
Overall, this should lead to improved cost efficiency, planning quality, communication,
standardization and knowledge capture and retention. [18].
The planning phases of the digital factory are based on the planning phases of factory
planning and can be divided according to Kettner, Schmidt, Greim [19] into target plan-
ning, preliminary planning, rough planning, real planning, detailed planning, execution
planning and execution. Tools such as process and material flow simulation can be used
to support the different planning scenarios [18].

3 Research Aim

The study examines how a digital twin of a production site can contribute to increased
value creation and how material flow simulations can be used as a suitable tool for this
purpose. Findings are required as the efficient design of the value chain is a key fac-
tor in ensuring the competitiveness of ETO companies [2, 3, 7]. The applicability of
 Towards the Digital Factory Twin in Engineer-to-Order Industries 83

Lean Management methods such as value stream mapping in SMEs and ETO indus-
tries has been discussed by several authors indicating high challenges for the respective
companies [20, 21]. Nevertheless, there are studies confirming the positive effect of the
given methods even for ETO industry and SMEs [20, 22]. Building digital competencies
and the existence of a simulation model enables fast evaluation of novel manufacturing
concepts, methods, and scheduling [6]. However, SMEs struggle with the evaluations
of such changes, as the impact is hard to foresee and does not justify the high effort
for implementation. Furthermore, SMEs lack the necessary knowledge, resources, and
time to adopt digital technologies [23]. This study shows, how novel principles and
methods of Lean Manufacturing for control cabinet manufacturing are more likely to
be accepted in conjunction with digital tools, specifically simulation. In this study, the
case of rescheduling and redesigning of existing factories is worked on to increase the
efficiency of the production processes. Thus, the study seeks to answer the following
research questions:
• RQ1: How can a digital factory twin contribute to increasing value creation and how
can a digital factory twin be implemented for ETO companies?
• RQ2: How can Lean Manufacturing principles, particularly the flow principle, be
applied in ETO industries to optimize the production process design?
Overall, this study seeks to combine the aspects of a systematic literature review
(SLR) and material flow simulation models to elaborate the maturity level of a simulation
model useable for replanning of factories and providing a possible research path to
DFT for ETO companies. Based on a SLR, the study outlines common aspects of the
DFT defined by several authors. Furthermore, the possibilities to use the DFT in ETO
production sites are elaborated and the correlating risks and barriers shown. In the second
part of the study a simulation is built for the control cabinet manufacturing industry.
Starting with an on-site process analysis, the current state is mapped using a material
flow simulation. Bottlenecks are identified, and lean manufacturing methods as the flow
principles2 are applied to test different action alternatives and determine the optimal
production process design.

4 Method

The method chosen for this study is divided into two parts: First, a SLR is performed,
which provides the opportunity to evaluate and synthesize the existing literature with
minimal bias, high efficiency, and consistency. Secondly, the study used a qualitative
research design, following the process of VDI-Guideline 3633 [25], conducting inter-
views with industry professionals from a company in the control cabinet industry, which
provided necessary data for a discrete event simulation.

2 According to Womack, Jones [24] the flow principle is one of the key principles of lean man-
agement and focuses on ensuring that the goods flow through the organization without any
interruptions or delays to eliminate waste in value creation.
84 M. Stoidner et al.

4.1 Systematic Literature Review

For the SLR, the study follows the methodology of Tranfield, et al. [26] to ensure
transparency, completeness, and reproducibility in the review process. The selection of
literature will be limited by several factors to ensure clear review paths [27, 28].
Timeline. In this literature review, the period under consideration is set between January
2010 and December 2021. The lower bound falls on the year 2010 because Shafto et al.
[29] from NASA defined the concept of DT for the first time in that year. The upper
bound of December 2021 is chosen to ensure that no changes are made in the sample
during the evaluation.
Databases. Only selected databases are searched for suitable literature. The databases
“Scopus”, “Web of Science” and “IEEE Xplore” are chosen.
Publication Types. Only peer-reviewed journal articles and conference proceedings
are included in the study.
Selection of Articles. A search string is developed with which relevant publications
are identified. The aspects “digital twin”, “factory” and “simulation” result from the
research question. It is important to consider related terms. Therefore, in addition to
the term “digital twin”, “digital thread” is included in the search string. There are also
synonyms for ‘factory’ such as ‘manufacturing’, ‘production’ and ‘operations’. These
terms are searched for in the title, abstract and keywords of the articles. This results in
the following search string:
ABS-TITLE-KEY ((“digital twin” OR “digital thread”) AND (factory OR manu-
facturing OR production OR operations) AND simulation).
As this study focusses on ETO companies, terms like engineering or ETO should be
included. The Authors concluded that adding this additional aspect to the search string
will reduce the amount of hits down to a minimum. Thus, this aspect is not added and
implications for ETO will be analyzed from the chosen literature.
Starting from the first 1644 hits, 1293 results remain after removing the duplicates.
By reading the titles and the summary, the selection was first reduced to 141 and further
to 38 results. After reviewing the entire text of each of the remaining hits and then
performing a snowball method, 30 results finally remained as the final basis for the
study. Excluded were papers that deal with DT outside the manufacturing sector, that
focus only on a single aspect of the factory, like logistics and studies that focused solely
on the DT of a product.
Classification. The authors independently conducted a comprehensive examination and
classification of all 30 results, carefully extracting pertinent subtopics, which are con-
cepts, properties, purposes, barriers and risks. Subsequently, the categorizations were
thoroughly discussed, leading to a consensus on the content. To mitigate within-study
bias and expectancy bias, the authors adhered to the guidelines and measures proposed
by Durach, Kembro, Wieland [30].
Evaluation. The authors engaged in detailed discussions and meticulous scrutiny of
the categorized articles to ensure coherence and identify potential contradictions. The
implications derived from the findings were critically evaluated to determine if they
 Towards the Digital Factory Twin in Engineer-to-Order Industries 85

were adequately substantiated or if further research was necessary for confirmation.

This review effectively addressed the research question presented in the initial chapter.

4.2 Material Flow Simulation

Secondly, a simulation project for a material flow simulation was conducted. The
simulation project follows the framework proposed in VDI-Guideline 3633 [25] .
The data collection is done with a process analysis at a German control cabinet
manufacturing company producing ETO. Through the analysis of a bill of materials
(BOM) for four projects, it was possible to classify the projects and the components for
the simulation. This is in line with the procedure of Braglia, Carmignani, Zammori [20]
who suggested this process for value stream mapping of ETO production. The resulting
classification is shown in Table 1.

Table 1. Bill of material for the recorded projects A-D

Components Project A Project B Project C Project D

Mounting plate 1 1 1 4
Cable duct 5 12 0 27
Mounting rail 2 7 0 17
Copper rail 0 1 5 1
Terminal 34 200 0 920
Wire bridges 4 10 0 80
Snap-on components 5 110 0 129
Screwed components 5 3 4 50
Wire connections 20 126 15 666
Ground connections 10 0 0 0
Additional parts (i.e. sensors…) 6 6 6 12

Classifying projects at the cabinet and component level ensures that quantities are
similar for other projects to generate realistic lead times for projects. To include the
uncertainties, ETO companies are facing, jobs are deployed in random order their count
of appearance summing up to the percentage of similar orders per year given by the
company. However, as the analysis focusses on the production processes and aims on
reducing bottlenecks in the production, the engineering of the projects were not integrated
in the material flow simulation. Finally, a discrete event simulation was performed in
Siemens PLM Plant Simulation [31] on the basis of the factory layout. However, the
process steps for control cabinet production given by literature are not sufficient for the
simulation, which is why the material flow was analyzed on-site. The resulting necessary
process steps are shown in Table 2. Using the BOM to calculate the process times resulted
in the process times per project shown in Table 3.
86 M. Stoidner et al.

A total of eight operators are involved in the simulated production of control cabinets.
Two operators are responsible for disassembly, mechanical processing, copper machin-
ing and pre-assembly. Five assembly operators are responsible for cutting cable ducts
and mounting rails, mechanical and electrical assembly, wire preparation, and wiring.
Apart from cutting, these operations are carried out sequentially at one assembly station.
One operator is testing the cabinets after production. The data points evaluated in the
simulation are operator capacity, output, and average lead times. The average lead time
ranges from 386.22 h for project C to 1006.97 h for project D. The remaining capacity of
the mechanical operators is between 71 and 55%, while the assembly operators have no
remaining capacity at all. An output of 63 units was calculated for project A, 32 units for
project B, 74 units for project C and 41 units for project D, which sums up to 210 units
and was accepted by the control cabinet manufacturing company as a realistic annual
output.

Table 2. Process steps and times in control cabinet manufacturing

Process step Time per unit in h

Incoming goods per cabinet 1
Disassembly per cabinet 0.5
Mechanical processing: Tapped hole 0.035
Mechanical processing: Cut-out 0.286
Processing of one copper rail 0.5
Cutting: cable duct 0.033
Cutting: mounting rail 0.042
Preassembly per mounting plate 2
Assembly of components: Snapping a component 0.024
Assembly of components: Screwing a component 0.12
Wiring and final assembly: complete process 8.08 – 99.49 (depending on project)
Testing 0.5 – 18 (depending on project)

Table 3. Resulting processing times per process and project

Process Steps Project A in h Project B in h Project C in h Project D in h

Incoming goods 1 1 1 2.5
Disassembly 0.5 0.5 0.5 1
Mechanical processing 1 1.75 1.5 2.5
Preassembly 2 2 2 8
Assembly of components 3.4 9.87 1.93 24
Wiring 14.01 40.87 8.05 99.49
Testing 3 10 0.5 18
 Towards the Digital Factory Twin in Engineer-to-Order Industries 87

5 Findings and Discussion

The first RQ is answered based on the literature. First, all 30 publications are assigned
to the classification topics they address, which can be seen in Table 4.

Table 4. Resulting classified topics and related publications

Topic Publications
Concepts [13, 14, 16, 32–41]
Properties [35, 42–54]
Purpose [32, 34, 36, 48, 52, 55]
Barriers and risks [23, 48, 56]

5.1 Analysis of the Literature

Concepts. As there are already many reviews like [11–16, 32, 33, 36, 41] about the
definition and different concepts of DT only the specific concepts presented for the DFT
are explained in this part.
The first concept introduced is the “Digital Surrogate”, which is defined by Shao,
Kibira [34] as an alternative to the DT. The Digital Surrogate is connected to the phys-
ical system through the Internet of Things (IoT) and intelligent sensors and actuators.
This allows the Digital Surrogate to be supplied with real-time data from the physical
system and to send control and action commands back to the physical system. The inte-
gration between model components within a Digital Surrogate is made possible through
interoperability standards.
Another presented concept is the “Digital Twin Shop-Floor”. This concept con-
nects the physical and virtual aspects of the shop-floor and ensures that both parts are
synchronized and optimize each other. According to Tao, Zhang [35] it consists of four
components: the physical shop-floor, virtual shop-floor, shop-floor service system, and
shop-floor digital twin data.
Gao, Lv, Hou, Liu, Xu [38] introduce the “Digital Twin Production Line”. This is
an intelligent production line system consisting of a physical production line, a digital
production line, an intelligent sensing system and an intelligent decision-making system.
It allows the management of production factors, production equipment, production data,
and production processes through correlation and integration.
Finally, the concept of a DFT that uses online simulation models to estimate model
parameters and keep a plant simulator in the same state as the actual plant or process is
discussed, which aligns with the idea of an “experimental DT” [37, 39, 40].
Properties A DT reflects the behavior of its physical counterpart throughout its entire
lifecycle. This concept can be applied in a variety of ways, such as recording and storing
comprehensive process data during manufacturing stages or using real-time information
about manufacturing challenges or component failures to pinpoint critical stages in the
88 M. Stoidner et al.

manufacturing process. In addition, predictive maintenance of machine tools can increase

machine availability and significantly reduce downtime [43].
To use a DFT in the context of Industry 4.0 and cyber-physical Systems (CPS),
scalability, modularity, and autonomy are relevant topics. The production system and
production units should be able to react autonomously to new orders, changed order
priorities, or disruptions in operation. Bi-directional data transmission should enable
autonomous decisions on action alternatives [44–47].
A DFT enables interoperability with a networked approach, as all models and data
are available in a consistent and coordinated environment, from models to data collection,
and from data analysis to data processing. Semantic descriptions are the key factor to
achieve this. In addition to CPS in production, their interaction with the products must
also be integrated [35, 44, 45, 48, 52–54].
Another characteristic is real-time model updating and synchronization. The
DFT offers a way to have reliable information on the state of the system, to detect
changes between the actual systems and the digital model within the DT, and to use the
information on changes to update the digital model [45, 49–51].
Purpose. The DFT supports the management of the production system by improv-
ing accuracy in management (i.e., planning and scheduling of jobs) and efficiency in
decision-making through detailed diagnoses and real-time monitoring. The DFT helps
monitoring and improving the production process. For example, simulations and opti-
mization applications for real-time monitoring and energy consumption monitoring can
improve overall plant efficiency, reduce costs, and increase productivity [32, 36, 48, 55].
A DFT can be used for the design of the manufacturing system, factory logistics, or
work environment. The production of the future will constantly need to change due to
various influencing factors, and every change must be planned [32, 36, 55].
In the area of maintenance, a DFT can also contribute since the effects of changes
in the condition of a production system on upstream and downstream processes can be
evaluated and predictive maintenance measures determined. Machine conditions can be
monitored based on descriptive methods and machine learning algorithms. In addition,
machine and process data from different phases of the machine lifecycle can be pro-
cessed more efficiently and a higher level of transparency can be achieved regarding the
condition of a machine [32, 36, 55].
Shao, Kibira [34] see potential in the reuse of models. The development and inte-
gration of model components or simulations of subsystems can be time-consuming,
expensive, and uncertain. In the future, the DFT should contribute to better and more
efficient reuse of models.
The resulting capabilities of DFT are diagnosis, monitoring and prediction, which
enables analyzing the past, predicting the future, and thus monitor, control, and optimize
current factory operations [48, 52].
Barriers and Risks. Several authors have identified current challenges and barriers
to use a digital twin in the manufacturing industry. Pires, Cachada, Barbosa, Moreira,
Leitao [48] are stating seven categories where the barriers stated by other authors can
be included. Perno, Hvam, Haug [23] have identified six categories of barriers that arise
during the development and implementation of DFZ. Uhlemann, Lehmann, Steinhilper
 Towards the Digital Factory Twin in Engineer-to-Order Industries 89

[56] have listed difficulties that can be attributed to the six categories identified by Perno,
Hvam, Haug.
IT Infrastructure. According to Pires, Cachada, Barbosa, Moreira, Leitao [48] the cur-
rent IT infrastructure presents a challenge for DT, just as it does for analytics and IoT.
To run effectively, a DT needs infrastructure that enables IoT and data analytics. In the
manufacturing industry, this infrastructure is rarely found and expensive to invest in [23,
56].
Useful Data. The data required for a DT needs to be of high quality, noise-free, and
come from a constant and uninterrupted stream. Inconsistent and poor data put the
DT at risk of underperforming. The number and quality of IoT signals are crucial to
ensuring that a DT performs efficiently. This barrier is exceptionally high for SMEs in
the manufacturing industry as IoT-ready sensors are seldom used [23, 48, 56].
Privacy and Security. DT poses a risk to sensitive system data and require proper privacy
and security considerations. The definition and compliance with security and privacy
regulations are crucial for enabling the [23, 48, 56].
Trust. Trust is a challenge for both end-users and organizations, and DT technology
needs to be explained at a foundational level to overcome this challenge. Model validation
is another way to ensure user trust in Digital Twins. Furthermore, bureaucratic efforts
and other organizational issues are hindering the reduction of trust issues among users
[23, 48, 56].
Expectations. More understanding is needed regarding the expectations of Digital
Twins. Appropriate action must be taken when developing DT systems, and it is essential
to discuss the positive and negative effects of using DT [23, 48, 56].
Modelling. The modelling of DT needs a standardized approach from initial design to
simulation to ensure domain and user understanding and information flow between each
stage of development and implementation. Additionally, to be able to model a DT a
high number of specialists and experts are needed. Transferring domain use informa-
tion to each development and functional stage of modelling ensures compatibility with
domains like IoT and data analytics, which is necessary for the successful use of DT
in the future. In addition, performance aspects must be considered. Manual collection
of motion data is still widespread but contradicts the necessary real-time availability in
most manufacturing use cases [23, 48, 56].
Discussion. Before RQ1 can be answered, a clear understanding of the concept of a DFT
must be provided. This can be done, synthesizing the results of the two classifications
for DFT concept and properties. These show that only a limited amount of literature is
dealing with DFT. Thus, research is not differentiating between a DT for a product or a
complete production site. However, the different concepts provided by literature indicate
the need for a clear understanding of the DFT, as similarities can be seen within the
concepts. The DFT must fulfil the criteria depicting the entire lifecycle, being scalable,
modular, and autonomous, creating interoperability and real-time model synchronization
as stated in the part “properties” of the literature synthesis. This is shown in Fig. 1, where
the common understanding of a DT as the combination of digital thread and digital model
90 M. Stoidner et al.

[12] is depicted. The combination of several digital twins of products can be called the
DT of a system and the DFT is the combination of several DT of a system.

Digital thread Digital Model

… … Digital Twin

… … Digital Twin
of Systems

Digital Factory
Twin

Fig. 1. Digital factory twin in the context of manufacturing

Synthesizing the information acquired from literature in the categories purposes and
barriers and risks allowed to answer RQ1, outlining possible value creation for ETO
companies. These are within the scope of diagnosis, monitoring, and prediction. Litera-
ture provided high challenges for restructuring ETO production layouts and evaluating
investment effects [7, 20]. The DFT will be a powerful tool to assess the current state of
the production, evaluate bottlenecks and predict changes to the ETO factory by rearrang-
ing processes. Product classifications can be used to enable the simulations [20, 37, 39,
40]. However, the necessary data as well as the modelling efforts pose high challenges
for ETO companies. This is due to the need for classification of products as well as the
lack of experts [23, 48, 56].

5.2 Material Flow Simulation

To answer RQ2, a simulation study based on the simulation model is conducted to visu-
alize waste in the production process. The optimum design of the production process
must be determined to increase the efficiency. To do so, the lean management princi-
ple of creating material flow is used. It must be considered that ETO products often
consist of expensive parts, are ordered for the specific project [57] and, therefore, tie
up a lot of capital. Consequently, a trade-off must be made between high output and
short lead time. The analysis of the initially introduced simulation model has shown a
bottleneck at the assembly stations, while mechanical operators are having remaining
capacity. This occurs when all five operators are busy in assembly and therefore the
employees in mechanical production must wait for buffer stocks to empty. To reduce the
processing time per step in the assembly the process is divided into electrical assembly
and wiring. Since the production steps of assembly and wiring still require the longest
machining times, the model should provide several parallel workstations for these pro-
cesses. This results in a reduction of the cycle time. Cutting should also be carried out
 Towards the Digital Factory Twin in Engineer-to-Order Industries 91

as an upstream process to assembly, so that the time required for this does not affect the
lead time for assembly. To be able to implement these measures, the layout of the factory
will be adapted and aligned according to the flow principle. The assembly and wiring
workstations are arranged in cells. When allocating the employees to the process steps,
care is taken to ensure that the workers must have appropriate qualifications for certain
activities and that the operators nevertheless have varied activities. Due to low personnel
utilization, mechanical production will in future be handled by only one person. One
operator will remain permanently in the testing department. The remaining six employ-
ees are divided among the other process steps as follows. One operator is responsible
only for electrical assembly, four workers carry out the wiring and one person is flexi-
bly deployed in both assembly and wiring. This allows the operators to rotate between
assembly and wiring at a certain rhythm, thus preventing monotony. Available produc-
tion space is limiting the number of creatable workstations or cells for assembly and
wiring, thus a maximum of eight stations can be implemented. Due to the high amount of
time necessary for the wiring process, the ideal combination of working stations is two
places for assembly and six places for wiring leading to the following results visualized
in Table 5. The remaining capacity of the mechanical operators is equal to the testing
operator at about 17%. A total output of 237 units is calculated. This is an increase in
output of 12.86% compared to the first model using the same number of operators in the
production process.
Utilizing a simulation model based on product classifications allows ETO companies
to assess rearrangements of their factory and identify bottlenecks within the production
site. The result of this study outlined important factors that need to be kept in mind when
replanning the production site.

Table 5. Summarized results of the simulation projects

Key Performance Indicator Project A Project B Project C Project D

Initial Avg. Lead Time 393.27 h 693.37 386.22 h 1006.97 h
Initial Output 63 32 74 41
Optimized Avg. Lead Time 259.77 h 436.42 h 210.57 h 913.77 h
Optimized Output 72 35 83 47
Increase of Output (per project) 9 / 14.29% 3 / 9.38% 9 / 12.16% 6 / 14.63%
Increase of Output (Sum) 27 / 12.86%

Digital mapping of manufacturing through material flow simulation is the first step in
the digital transformation of ETO manufacturers towards a DFT. This enables the control
cabinet manufacturer to optimize workflows and improve the utilization of resources.
With the current state of material flow simulation, companies can plan changes in their
production, e.g., when purchasing a machine in wire harnessing or cutting in advance.
Furthermore, the factors explained within this study enhance existing frameworks for
process improvements in ETO companies [20, 21].
92 M. Stoidner et al.

6 Conclusion
The control cabinet manufacturing industry faces major challenges, such as inefficient
processes in value creation, digital transformation as well as ETO specific challenges
like lack of feasible automation and media breaks [1, 2, 4, 17].
This work contributes to the fulfilment of the need for action by providing concepts
of flow production for control cabinet manufacturing to make the process chain more
efficient and to increase productivity as in the example an increase of 12.86% in out-
put was achieved. In addition, a material flow simulation was found to be a suitable
digital tool for uncovering bottlenecks and planning optimizations for control cabinet
manufacturing. Furthermore, the classification on a cabinet and component basis is a
suitable method for control cabinet manufacturers to decide on the right manufacturing
order to avoid bottlenecks and waste. The study contributes to the research field DFT
by providing applications as well as characteristics of a DT of a complete production
site. Additionally, this work contributes to the ETO research field as it enhances exist-
ing frameworks to implement process improvements by outlining important factors and
explaining the creation of a simulation model. However, to be able to call the resulting
simulation model a DFT extent research is necessary for the control cabinet industry.
Practitioners and managers in ETO companies can use the findings to establish an in-
depth understanding and basis for the implementation of DFT and simulation for a digital
transformation.
This study shows current barriers to creating a DT of a production system, which
is especially significant for control cabinet manufacturing companies. These include
retrofit data acquisition at existing manual or semi-automatic process stations, lack of
specialists and media interruptions [23, 48, 56]. This leads to a further need of research
on how a material flow simulation model depicting the current factory layout and process
control can be created by persons without prior knowledge in simulations. This will also
address the lack of expertise in auxiliary processes like production logistics. Further
research in creating a DFT for ETO manufacturers is enabled with this study, as the core
properties of a DFT, which include usage along the entire lifecycle, scalability, autonomy,
modularity, interoperability as well as real-time model updating and synchronization,
were identified.

References
1. Großmann, C., Graeser, O., Schreiber, A.: ClipX: on the way to industrialization of control
cabinet construction (in German). In: Vogel-Heuser, Bauernhansl et al. (Hg.) 2017 – Handbook
Industry 4.0 Vol.2 (in German), pp. 169–187 (2017)
2. Tempel, P., Eger, F., Lechler, A., et al.: Control Cabinet Manufacturing 4.0: a study on the
automation and digitization potentials in the production of control cabinets and switchgear
in the classic mechanical and plant engineering sector (in German), Stuttgart (2017)
3. Olhager, J.: Strategic positioning of the order penetration point. Int. J. Prod. Econ. 85, 319–329
(2003). https://doi.org/10.1016/S0925-5273(03)00119-1
4. Willner, O., Powell, D., Duchi, A., et al.: Globally distributed engineering processes: making
the distinction between engineer-to-order and make-to-order. In: ElMaraghy, H. (ed.) Pro-
ceedings of the 47th CIRP Conference on Manufacturing Systems, 17th edn., vol 17. Elsevier
B.V., pp. 663–668 (2014)
 Towards the Digital Factory Twin in Engineer-to-Order Industries 93

5. Joppen, R., von Enzberg, S., Kühn, A., et al.: Investment decisions against the background of
digitalization using the example of control cabinet construction (in German). Zeitschrift für
wirtschaftlichen Fabrikbetrieb 114, 483–487 (2019). https://doi.org/10.3139/104.112119
6. Riley, L.A.: Discrete-event simulation optimization: a review of past approaches and propo-
sitions for future direction. In: Abhari, A. (ed.) Proceedings of the 2013 Summer Simulation
Multiconference: Summer Computer Simulation Conference and Work in Progress (SCSC
2013 and WIP 2013), vol. 47, pp. 1–8. Curran Associates, Inc, Red Hook, NY (2013)
7. Bertrand, J., Muntslag, D.R.: Production control in engineer-to-order firms. Int. J. Prod. Econ.
30–31, 3–22 (1993). https://doi.org/10.1016/0925-5273(93)90077-X
8. Olhager, J.: The role of the customer order decoupling point in production and supply
chain management. Comput. Ind. 61, 863–868 (2010). https://doi.org/10.1016/j.compind.
2010.07.011
9. Wikner, J., Rudberg, M.: Integrating production and engineering perspectives on the customer
order decoupling point. Int. J. Oper. Prod. Manag. 25, 623–641 (2005). https://doi.org/10.
1108/01443570510605072
10. Wijngaard, J., Wortmann, J.C.: MRP and inventories. Eur. J. Oper. Res. 20, 281–293 (1985).
https://doi.org/10.1016/0377-2217(85)90001-3
11. Sjarov, M., Lechler, T., Fuchs, J., et al.: The digital twin concept in industry – a review and
systematization. In: 2020 25th IEEE International Conference on Emerging Technologies and
Factory Automation (ETFA), pp. 1789–1796. IEEE (2020)
12. Stark, R., Anderl, R., Thoben, K.-D., et al.: WiGep position paper on the topic of “Digital
Twin” (in German) (2020) . https://secureservercdn.net/160.153.137.184/b7s.1f6.myftpu
pload.com/wp-content/uploads/2022/05/Final_WiGeP_Positionspapier_Digital_Twin.pdf.
Accessed 27 Sep 2022
13. Resman, M., Protner, J., Simic, M., et al.: A five-step approach to planning data-driven digital
twins for discrete manufacturing systems. Appl. Sci. 11, 3639 (2021). https://doi.org/10.3390/
app11083639
14. Guerra-Zubiaga, D., Kuts, V., Mahmood, K., et al.: An approach to develop a digital twin for
industry 4.0 systems: manufacturing automation case studies. Int. J. Comput. Integr. Manuf.
34, 933–949 (2021). https://doi.org/10.1080/0951192X.2021.1946857
15. Barricelli, B.R., Casiraghi, E., Fogli, D.: A survey on digital twin: definitions, characteristics,
applications, and design implications. IEEE Access 7, 167653–167671 (2019). https://doi.
org/10.1109/ACCESS.2019.2953499
16. Assad Neto, A., Da Ribeiro, S.E., Deschamps, F., et al.: Digital twins in manufacturing: an
assessment of key features. Procedia CIRP 97, 178–183 (2021). https://doi.org/10.1016/j.pro
cir.2020.05.222
17. Rasheed, A., San, O., Kvamsdal, T.: Digital twin: values, challenges and enablers from a
modeling perspective. IEEE Access 8, 21980–22012 (2020). https://doi.org/10.1109/ACC
ESS.2020.2970143
18. Verein Deutscher Ingenieure e.V. (2008) Digital Factory - Basics (in German) ICS 03.100.50,
35.240.50(4499 Sheet 1). Accessed 04 Apr 2022
19. Kettner, H., Schmidt, J., Greim, H.-R.: Guide to systematic factory planning with numerous
checklists (in German). Hanser, München (2010)
20. Braglia, M., Carmignani, G., Zammori, F.: A new value stream mapping approach for complex
production systems. Int. J. Prod. Res. 44, 3929–3952 (2006). https://doi.org/10.1080/002075
40600690545
21. Matt, D.T.: Adaptation of the value stream mapping approach to the design of lean engineer-
to-order production systems. J. Manuf. Technol. Manag. 25, 334–350 (2014). https://doi.org/
10.1108/JMTM-05-2012-0054
22. Bhamu, J., Singh Sangwan, K.: Lean manufacturing: literature review and research issues.
Int. J. Oper. Prod. Manag. 34, 876–940 (2014). https://doi.org/10.1108/IJOPM-08-2012-0315
94 M. Stoidner et al.

23. Perno, M., Hvam, L., Haug, A.: Enablers and Barriers to the Implementation of Digital
Twins in the Process Industry: A Systematic Literature Review. In: 2020 IEEE International
Conference on Industrial Engineering and Engineering Management (IEEM), pp. 959–964.
IEEE (2020)
24. Womack, J.P., Jones, D.T.: Lean thinking—banish waste and create wealth in your corporation.
J. Oper. Res. Soc. 48, 1148 (1997). https://doi.org/10.1057/palgrave.jors.2600967
25. Verein Deutscher Ingenieure e.V. (2014) Simulation of logistics, material flow and production
systems - basics (in German) ICS 03.100.10(3633 Sheet 1). Accessed 05 May 2022
26. Tranfield, D., Denyer, D., Smart, P.: Towards a methodology for developing evidence-
informed management knowledge by means of systematic review. Br. J. Manage. 14, 207–222
(2003). https://doi.org/10.1111/1467-8551.00375
27. Rojon, C., Okupe, A., McDowall, A.: Utilization and development of systematic reviews in
management research: What do we know and where do we go from here? Int. J. Manag. Rev.
23, 191–223 (2021). https://doi.org/10.1111/ijmr.12245
28. Mulrow, C.D.: Rationale for systematic reviews. BMJ 309, 597–599 (1994). https://doi.org/
10.1136/bmj.309.6954.597
29. Shafto, M., Conroy, M., Doyle, R., et al.: DRAFT Modeling, Simulation, Information
Technology & Processing Roadmap: Technology Area 11 (2010)
30. Durach, C.F., Kembro, J., Wieland, A.: A new paradigm for systematic literature reviews
in supply chain management. J. Supply Chain Manag. 53, 67–85 (2017). https://doi.org/10.
1111/jscm.12145
31. o. V. (2008) Siemens Data Sheet. https://plant-simulation.de/wp-content/uploads/Plant_Sim
ulation_Produkt%C3%BCbersicht.pdf. Accessed 23 Sep 2022
32. Kritzinger, W., Karner, M., Traar, G., et al.: Digital Twin in manufacturing: a categorical
literature review and classification. IFAC-PapersOnLine 51, 1016–1022 (2018). https://doi.
org/10.1016/j.ifacol.2018.08.474
33. Redelinghuys, A.J.H., Basson, A.H., Kruger, K.: A six-layer architecture for the digital twin:
a manufacturing case study implementation. J. Intell. Manuf. 31, 1383–1402 (2020). https://
doi.org/10.1007/s10845-019-01516-6
34. Shao, G., Kibira, D.: Digital manufacturing: requirements and challenges for implementing
digital surrogates. In: 2018 Winter Simulation Conference (WSC), pp. 1226–1237. IEEE
(2018)
35. Tao, F., Zhang, M.: Digital twin shop-floor: a new shop-floor paradigm towards smart manu-
facturing. IEEE Access 5, 20418–20427 (2017). https://doi.org/10.1109/ACCESS.2017.275
6069
36. Park, K.T., Yang, J., Noh, S.D.: VREDI: virtual representation for a digital twin application
in a work-center-level asset administration shell. J. Intell. Manuf. 32, 501–544 (2021). https://
doi.org/10.1007/s10845-020-01586-x
37. Schluse, M., Priggemeyer, M., Atorf, L., et al.: Experimentable digital twins—streamlining
simulation-based systems engineering for industry 4.0. IEEE Trans. Ind. Inf. 14, 1722–1731
(2018). https://doi.org/10.1109/TII.2018.2804917
38. Gao, Y., Lv, H., Hou, Y., et al.: Real-time modeling and simulation method of digital twin
production line. In: 2019 IEEE 8th Joint International Information Technology and Artificial
Intelligence Conference (ITAIC), pp. 1639–1642. IEEE (2019)
39. Schluse, M., Rossmann, J.: From simulation to experimentable digital twins: Simulation-
based development and operation of complex technical systems. In: 2016 IEEE International
Symposium on Systems Engineering (ISSE), pp. 1–6. IEEE (2016)
40. Martinez, G.S., Sierla, S., Karhela, T., et al.: Automatic generation of a simulation-based
digital twin of an industrial process plant. In: IECON 2018 - 44th Annual Conference of the
IEEE Industrial Electronics Society. IEEE, pp. 3084–3089 (2018)
 Towards the Digital Factory Twin in Engineer-to-Order Industries 95

41. Bazaz, S.M., Lohtander, M., Varis, J.: 5-Dimensional definition for a manufacturing digital
twin. Procedia Manuf. 38, 1705–1712 (2019). https://doi.org/10.1016/j.promfg.2020.01.107
42. Weyer, S., Meyer, T., Ohmer, M., et al.: Future modeling and simulation of CPS-based fac-
tories: an example from the automotive industry. IFAC-PapersOnLine 49, 97–102 (2016).
https://doi.org/10.1016/j.ifacol.2016.12.168
43. Pang, T.Y., Pelaez Restrepo, J.D., Cheng, B., et al.: Developing a Digital Twin and Digital
Thread Framework for an ‘Industry 4.0’ Shipyard (2020)
44. Rosen, R., von Wichert, G., Lo, G., et al.: About the importance of autonomy and digital
twins for the future of manufacturing. IFAC-PapersOnLine 48, 567–572 (2015). https://doi.
org/10.1016/j.ifacol.2015.06.141
45. Um, J., Weyer, S., Quint, F.: Plug-and-simulate within modular assembly line enabled by
digital twins and the use of AutomationML. IFAC-PapersOnLine 50, 15904–15909 (2017).
https://doi.org/10.1016/j.ifacol.2017.08.2360
46. Yildiz, E., Møller, C., Bilberg, A.: Virtual Factory: Digital Twin Based Integrated Factory
Simulations. In: Gao, R.X., Ehmann, K. (eds.) 53rd CIRP Conference on Manufacturing
Systems 2020, vol. 93. Elsevier B.V., pp. 216–221 (2020)
47. Jeon, S.M., Schuesslbauer, S.: Digital Twin Application for Production Optimization. In:
2020 IEEE International Conference on Industrial Engineering and Engineering Management
(IEEM), pp 542–545. IEEE (2020)
48. Pires, F., Cachada, A., Barbosa, J., et al.: Digital Twin in Industry 4.0: Technologies, Applica-
tions and Challenges. In: 2019 IEEE 17th International Conference on Industrial Informatics
(INDIN). IEEE, pp. 721–726 (2019)
49. Talkhestani, B.A., Jazdi, N., Schloegl, W., et al.: Consistency check to synchronize the Digital
Twin of manufacturing automation based on anchor points. Procedia CIRP 72, 159–164
(2018). https://doi.org/10.1016/j.procir.2018.03.166
50. Zipper, H., Diedrich, C.: Synchronization of industrial plant and digital twin. In: 2019 24th
IEEE International Conference on Emerging Technologies and Factory Automation (ETFA),
pp. 1678–1681. IEEE (2019)
51. Sommer, M., Stjepandić, J., Stobrawa, S., et al.: Automatic generation of digital twin based
on scanning and object recognition. In: Hiekata, K., Moser, B.R., Inoue, M., et al. (eds.)
Transdisciplinary Engineering for Complex Socio-technical Systems. IOS Press (2019)
52. Glatt, M., Sinnwell, C., Yi, L., et al.: Modeling and implementation of a digital twin of material
flows based on physics simulation. J. Manuf. Syst. 58, 231–245 (2021). https://doi.org/10.
1016/j.jmsy.2020.04.015
53. Cheng, Y., Zhang, Y., Ji, P., et al.: Cyber-physical integration for moving digital factories
forward towards smart manufacturing: a survey. Int. J. Adv. Manuf. Technol. 97, 1209–1221
(2018). https://doi.org/10.1007/s00170-018-2001-2
54. Jaensch, F., Csiszar, A., Scheifele, C., et al.: Digital twins of manufacturing systems as a base
for machine learning. In: 2018 25th International Conference on Mechatronics and Machine
Vision in Practice (M2VIP), pp. 1–6. IEEE (2018)
55. Cimino, C., Negri, E., Fumagalli, L.: Review of digital twin applications in manufacturing.
Comput. Ind. 113, 103130 (2019). https://doi.org/10.1016/j.compind.2019.103130
56. Uhlemann, T.H.-J., Lehmann, C., Steinhilper, R.: The digital twin: realizing the cyber-physical
production system for industry 4.0. Procedia CIRP 61, 335–340 (2017). https://doi.org/10.
1016/j.procir.2016.11.152
57. Adrodegari, F., Bacchetti, A., Pinto, R., et al.: Engineer-to-order (ETO) production planning
and control: an empirical framework for machinery-building companies. Prod. Plann. Control
26, 910–932 (2015). https://doi.org/10.1080/09537287.2014.1001808
 Has the Pendulum Swinged Too Far
from Just-in-Time to Just-in-Case
in the Aftermaths of Covid-19?

Jenny Bäckstrand1(B) and Andréas Malmstedt2

1 School of Engineering, Jönköping University, Gjuterigatan 5,
55111 Jönköping, Sweden
[email protected]
2 Ericsson AB, Kista, Stockholm, Sweden

Abstract. The last three years have been a challenge for many businesses and
have emphasized the need for resilience. An important leaver to achieve resilience
is inventory management. Having the right level of inventory has proven to be
extremely difficult, especially for semiconductors with their long process lead-
times and scarcity. Many companies have transitioned from just-in-time to just-
in-case supply chains and from being faced with under-absorption due to lack of
material to being faced with over-supply and over-stock situation when the demand
eased up. The lack of an integrated view of customer demand and supply changes
can lead to the bullwhip effect where uncertainty at every tier of supply can lead to
inflated stock levels for each successive upstream tier. Consequently, the business
implications of the fast development from undersupply to oversupply is widely
seen in the quarterly reports for 2022Q3 and in the profit warnings for 2022Q4
where companies explained the reduced profits due to inventory reduction in the
retail distribution. The purpose of this paper is to investigate the decision making
during the pandemic, and what changes companies have implemented to achieve
more resilient supply chains for semiconductors. Did the pendulum swing too
extreme from just-in-time to just-in-case? This is answered by a literature review
and an exploratory survey among Swedish manufacturing companies. The result
is the identification of challenges for Swedish companies in achieving sustainable
resilient supply chains instead of mitigating the disruptions.

Keywords: Supply Chains · Responsiveness · AI · Inventory Management ·

Semiconductors

1 Introduction
Many industries have during the last years been brought to their knees by supply scarcity
of semiconductors [1]. Having the right level of inventory has proven to be extremely
difficult, especially for semiconductors with their long process lead-times and scarcity.
In 2021, several automakers had to cancel their production plans due to shortage of
components [2]. Consumers were faced with “Out of Stock” signs in many stores of
© IFIP International Federation for Information Processing 2023
Published by Springer Nature Switzerland AG 2023
E. Alfnes et al. (Eds.): APMS 2023, IFIP AICT 691, pp. 96–111, 2023.
https://doi.org/10.1007/978-3-031-43670-3_7
 Has the Pendulum Swinged Too Far from Just-in-Time 97

consumer electronics or by rather long waiting times for new products. Apple’s iPhone 12
Pro came with up to 3 weeks waiting time when it was launched [2]. Another smartphone
maker, Xiaomi described the situation as “very, very serious situation with shortages
affecting the whole industry” [2]. Home appliance makers were also struggling to meet
demand. For example, Electrolux described its supply chain during 2021 as “strained”
in many areas, especially for electronic components [2]. Or as described it in the book
Chip War, people around the world began to understand how much their lives depend
on semiconductors [3].
The components to this ‘perfect storm’ are plentiful. During the covid pandemic
when many companies sent their staff home for remote work, the demand for PCs,
smartphones, and data centers skyrocketed together with home entertainment equipment,
here referred to as cloud and communication equipment, which all in all created a huge
demand for semiconductors.
This demand surge was placed on an already constrained supply chain due to aggres-
sive stockpiling from Chinese companies due to geopolitical tensions between the US
and China [2].
The automotive industry, for example, expected the sales to decline during the initial
phase of the Covid pandemic and cut chip orders to offload their expenses. When the
demand quickly recovered, the big semiconductor vendors had already reallocated the
capacity from the automotive industry to cloud and communication equipment [3].
The situation was worsened by the Covid lockdowns in Malaysia which impacted
the final assembly and packaging operations for many semiconductor manufacturers
who have outsourced these steps of the supply chain. In addition to reduced production,
the challenges of global logistics also contributed to the long lead-times and shortage
of semiconductors as well as wavers and other material needed for production of the
electronic components [4].
To add to an already strained situation, natural disruptions such the winter storm in
Texas 2021, which led to outage of gas, effected the semiconductor factories, e.g., wafer
fabs for the manufacturers NXP and Samsung which were forced to halt production
and all their work-in-process products were destroyed due to the quick shutdown. The
production for NXP was not resumed until 24 days later resulting in a huge lack in
the supply plan for several industries such as telecom, smartphones and automotive.
Another manufacturer of semiconductors to the automotive industry, Renesas, suffered
from a factory fire in April 2021 which led to a plant shut down and a reduced output of
10%. The factory plant was back to normal capacity during June 2021. To put all these
disruptions into context, the telecom, smartphone, and automotive industry had before
this strived for just-in-time deliveries and lean/minimal inventories to reduced tied up
capital.
However, in 2021 the production of chips was 13% higher than the year 2020. Hence,
Chris Miller [3] state that the semiconductor shortage is mostly a story of demand growth
rather than a supply issue. For the automotive industry, which suffered a several hundred
billion dollar hit to their revenue, there is thus reasons to re-think how they manage their
supply chains. According to [1], What goes for automotive companies, are also valid for
many other manufacturing companies.
98 J. Bäckstrand and A. Malmstedt

The chip industry has however pivoted hard from a record revenue to cost cutting as
it adjusts to a slump for semiconductors [5] which has been reported several semicon-
ductor companies during 2022Q4-2023Q1. This was first seen at the memory providers
Samsung and Micron who are seen as a bellwether for the global tech world [6]. Thus,
chip companies, including Intel, have pared back production plans and reduced capital
spending [7].
The world’s top memory-chip makers are now struggling with the deepest indus-
try downturn in more than a decade [8]. As a concrete example Micron saw its fiscal
second-quarter revenue plunged 53% from a year earlier. That follows a 47% drop in the
previous quarter. Economic conditions and damped consumer demand, as well as inven-
tory adjustments from its customers, led to the drop in profit [6], or as the CEO of Intel
said: “The Largest Ever Inventory Correction By Customers In Q1” [7]. Akash Palkhi-
wala, Qualcomm’s chief financial officer [5] concludes “It’s an unprecedented change
over a short period of time.” He further concluded that: “The strength of the boom is
turning into extra pain in the current downturn. When consumers clamored for digital
goods, many manufacturers responded by stocking up on chips to respond to demand
more quickly. Now that consumers aren’t buying as many phones or PCs, manufacturers
are running through those chip inventories rather than placing new orders”.
Consequently, many companies have transitioned from just-in-time to just-in-case
supply chains and from being faced with under-absorption due to lack of material to being
faced with over-supply and over-stock situation when the demand eased up. The lack of
an integrated view of customer demand and supply changes can lead to the “bullwhip
effect” where uncertainty at every tier of supply can lead to inflated stock levels for
each successive upstream tier. The business implications of the fast development from
undersupply to oversupply is widely seen in the quarterly reports for 2022Q3 and in
the profit warnings for 2022Q4 where companies explained the reduced profits due to
inventory reduction in the retail distribution.
The purpose of this paper is to investigate the decision making during the pandemic,
and to analyze if the pendulum swung too extreme from just-in-time to just-in-case?
This is done by answering the following research question:
RQ1: what changes have companies implemented to achieve more resilient supply
chains for semiconductors?
This is answered by a literature review and an exploratory survey among Swedish
manufacturing companies. The result is the identification of challenges for Swedish
companies in achieving sustainable resilient supply chains instead of mitigating the
disruptions.

2 Theory

This section initially covers the areas of supply chain risks and risk management where
building blocks for a resilient supply chain are presented. A resilient supply chain needs
to be able to hance supply and demand fluctuations without amplifying them, the next
section consequently covers the bullwhip effect and the additional complication when
a bottleneck is present. Finally, the semiconductor manufacturing and supply chain are
introduced.
 Has the Pendulum Swinged Too Far from Just-in-Time 99

2.1 Supply Chain Resilience and Risk Management

Supply chain risks can, as described earlier, cause severe supply-chain problems, causing
unanticipated changes in flow due to disruptions or delays. This may result in losses of
revenue and incur high recovery costs [4]. Due to its nature, the frequency of disruptions
varies as well as the magnitude of the problem in size and duration.
Organizations that manage these risk events better than others share one common
trait: resilience, according to Sheffi [9]. Supply chain resilience is defined as:
“The capacity of a system (supply chain) to return to its original state or move to a
new, more desirable state after being disturbed” [9].
Hence the question arises How do you build a resilient supply chain? When reviewing
the literature in this field a couple of common factors emerge which is summarized in
below in Table 1 [10].

Table 1. Building blocks for a resilient supply chain.

Building blocks for a resilient Translated into Factors References

supply chain
Supply Chain Risk Mapping Country of origin, back-end Christopher & Peck (2004)
and front-end locations, [11],
lead-times Chopra & Sodhi (2004), [12]
Lund et al., (2020) [13]
Inventory & Capacity Buffer Buffer size, buffer target, Chopra & Sodhi (2004) [12],
capacity Lund et al., (2020) [13],
Sheffi (2005) [9]
Multisource Number of suppliers Christopher & Peck (2004)
[11],
Chopra & Sodhi (2014) [12],
Lund et al., (2020) [13], Sheffi
(2005) [9], Shih (2020) [14]
Containment Number of products, Volume Chopra & Sodhi (2014) [12]
per region/customer, Number
of suppliers per region
Modular design Volume per module Lund et al., (2020) [13]
Supply chain control tower Locations of disruptions, Goering et al., (2018) [15]
frequency of disruptions, Ivanov & Dolgui (2020) [16]
duration of disruptions, size of Linton & Vakil (2020) [17]
disruptions, supply visibility

In the article ‘The Semiconductor Crisis Should Change Your Long Term Supply
Chain Strategy’ [1], the authors argue that supply chain managers should use the play
book of Big Tech companies to create a competitive advantage, innovation and risk
management. From the learnings from the component crises during 2017, actors like
Apple and dell immediately acted and the main learnings from that playbook is presented
100 J. Bäckstrand and A. Malmstedt

below in Table 2 where the main learnings are compared to the supply chain building
blocks presented in Table 1.

Table 2. Supply chain building blocks compared to main learnings.

Building block Learning

Supply chain control tower Establishment of control of the Bill of material
Inventory and Capacity buffer Firm order or capacity commitment
Inventory and Capacity buffer Ensure their supplier earmark specific component for their sole
use
Supply chain control tower Collaborate with suppliers to track and trace every order
Modular design Reinforce supplier collaboration including R&D

2.2 Bullwhip Effect and Bottle Necks

The bullwhip effect occurs when the demand order variabilities in the supply chain are
amplified as they moved up the supply chain [18]. Lee et al. [18] identified four major
causes of the bullwhip effect: Demand forecast updating, Order batching, Price fluctua-
tion, and Rationing and shortage gaming. In the situation during the Covid pandemic, we
can see that all four of these causes have been present. The resulting amplified demand,
consequently led to semiconductors becoming a bottleneck item [19, 20].
In a situation with limited manufacturing capacity, the position in the supply chain
with highest demand density can also be termed a bottleneck, but in order to differentiate
between the two, we will use the term choke point when referring to the supply chain.
The theory of constraints [21] states that any system contains a choke point that prevents
it from achieving its goals.

2.3 Semiconductor Manufacturing and Supply Chain

What is a semiconductor?
Semiconductors play an important role in all electronic devices such as smartphones,
computers, and appliances but also in automotive and telecom. A semiconductor is a piece
of material, usually silicon, which has electrical conductivity which value vary between
a conductor and an insulator given some conditions. This makes it an ideal medium
to control electrical current in electrical devices [22]. The semiconductor material has
millions of microscopic transistors carved into the material.
Semiconductor components are built up in a series of nanofabrication processes
performed on the surface of substrates made from highly pure single crystal silicon
[23].It is a multiple-step photolithographic and physico-chemical process in which the
electrical devices gradually are being created on a substrate, known as wafers. Wafer
sizes are ranging from 5 nm for the most complex use cases, up to 300 mm, where 300
mm is the most common wafer size [4].
 Has the Pendulum Swinged Too Far from Just-in-Time 101

Semiconductors are normally divided into three distinct categories [3]: Logic: proces-
sors used in smartphones and computers, Memory: used short-term and long-term mem-
ory for storing data, and Chips: integrated circuit – analog like sensors, radio frequency
and semiconductor chip.

Manufacturing
Semiconductor manufacturing is normally divided into design, front-end, and back-end
processes, see Fig. 1 [24]. The manufacturing of the wafer is being done in a foundry
called “fabs” where the main part of the factory is the clean room. The process lead-time
for a 300-mm wafer is 15 weeks. The manufacturing is fully automated and carried out
in dedicated equipment. The equipment used in these processes is very complex and
expensive to invest in.

Fig. 1. The flow of semiconductor manufacturing [24]

The manufacturing equipment is only produced by two vendors, ASML and Nipon
Industries [3]. Building a new wafer fab requires an investment of > 1 Billon dollars
and several years of lead time including the source and installment of the process equip-
ment. Given the high investment requirements and extensive R&D, very few companies
have the capability to design and produce semiconductors in this $500bn industry [25].
Furthermore, Moore’s Law predicted that the number of transistors that you could fit
on a chip would double every two years and currently the most complicated chips can
contain more than 1 billion transistors [26]. Hence the rapid innovation is expected to
continue.
According to Chris Miller [3] “Fabrication and miniaturizing semiconductors have
the greatest engineering challenge of our time”. One of the most advanced firms of semi-
conductor manufacturing is Taiwan Semiconductor Manufacturing Company, TSMC.
In its most advanced fabrication plant, Fab18, and in its sister plants, almost all of the
worlds most advanced chips are build [3], consequently we argue that this is one of the
most important choke points in the global supply chain.

Semiconductor Supply Chain

The global semiconductor supply chain is illustrated in Fig. 2 [27].
After the wafers are completed in the front-end process, the flow continues to another
factory, in many cases in Asia, where back-end process, the assembly and packaging of
the component takes place. In this process, wafer cutting is performed to separate the
semiconductors and complete the semiconductor. The cut semiconductors are fixed, and
terminals are attached or covered with resin. As illustrated in Fig. 2, the supply chain of
semiconductors is divided into several phases from design and chip development, into
102 J. Bäckstrand and A. Malmstedt

Fig. 2. Semiconductor supply chain

chip production, assembly, and distribution. Design is mainly done is US, manufacturing
in US, Taiwan, and South Korea whereas assembly and packaging are done in Malaysia
and China [3]. Due to the highly complex processes and capital employed, each stage
in the supply chain requires specialization in order to achieve efficiency of scale. One
form of specialization is fabless companies which only focus on chip design such as
NVIDA and Qualcomm who outsource the operations to Intel or TSMC, or work with
partnership [28]. Due to this, several actors are involved in the semiconductor supply
chain.
The customer order point is normally between front end and back end in stored
wafers. Until the point wafers are produced make-to-stock in high batches. The final
assembly is make-to-order. As a consequence of its lengthy processes, it normally takes
8–15 months from the order is placed until delivery [20]. Additionally, the semiconductor
industry’s upstream position in the overall supply chain amplifies the bullwhip effect
[20]. The long lead time together with the many actors involved adds on another layer
on the bullwhip. Hence intense coordination and orchestration is required to manage the
supply chain successfully.
In summary, semiconductor manufacturing is a very complex process that requires
high capital investments and R&D efforts. The supply chain for semiconductors is long
with long processes, several nodes, actors and with intense transportation. Reviewing
the supply chain map, one clearly identifies a few vital coke points.

3 Method

In order to fulfill the purpose and answer the research questions we have combined a
literature study with an interview study. An analysis and summary of literature have
guided the interview guide used for the structured interviews. The unit of analysis that
has been studied is the case companies’ actions in relation to the semiconductor supply
chain. The theoretical basis is built on supply chain resilience, the bullwhip effect in
supply chains and an insight into how semiconductors are manufactured and procured.
 Has the Pendulum Swinged Too Far from Just-in-Time 103

3.1 Sample
The interviewees were selected to represent companies from different industries, of
different sizes and with different semiconductor density. With density we are referring
to the number of semiconductors in relation to the total Bill-of-Materials (BOM) and
is measured in semiconductor/electronic components part of the total sourcing spend.
In addition to interviewing manufacturing companies, we also conducted an interview
with one of the main providers of logistics services in the semiconductor industry, here
referred to as Company Z. The target was to compare and to validate the observations
from the manufacturing companies and to receive a consolidated view since the company
data set only gives an indication and not a complete view. An overview of the companies
interviewed in this study can be found in Table 3.

3.2 Interview Guide

The interview guide started with five free text questions to get an understanding of the
business situation at the case company.
1. How big part of your sourcing spend is in semiconductors/electronic components?
2. How are the components sourced?
a. from first tier (%)
b. from electronic distributors (%)
c. from the open market (%)
3. How have your material cost changed since 2019 for electronics? (%)
4. How have your inventory turnover changed since 2019?
5. How have your revenue changed since 2019?

Table 3. Overview of case companies.

Company A B C D E
Revenue 5 BSEK 86 BSEK 6 BSEK 6 BSEK 13 BSEK
Industry Automotive Telekom Defence Shop concepts Power
Q1 70% 25% 7% 5–10% 3%
Q2
First tier 95% 98% 20% Yes 100%
Distributor 5% 70% Yes
Open market 0 10% Yes

This was followed by seven questions with four to five predetermined answers, plus
the alternative “other” as a free text answer, see below for a comprehensive list (please
observe that the alternative “other” only is included in tables 4–10 when this alternative
was selected). The themes of these questions are inspired from Table 1, with the addition
of a question regarding supply chain governance to cover how the collaborate at manage
their supply chain.
104 J. Bäckstrand and A. Malmstedt

6. To which extent has your ability to deliver been affected by the general shortage of
electronics?
a. We have been able to supply all demand at normal lead times with no changes in
our supply chain.
b. We have been able to supply all demand at normal lead times with changes in
our supply chain.
c. We have been able to supply all demand with prolonged lead time.
d. Our capability to deliver has been disturbed.
7. This section concerns supply chain control tower and your knowledge about other
actors within semiconductors in your supply chain and availability of data.
a. We know the original source from where our material is sourced from.
b. We are well aware where our suppliers are located.
c. We always have information on inventory availability, lead times and delivery
dates in our material supply.
d. The location and status of some of our material is always visible throughout the
distribution network (e.g., distribution centers, transportation)
e. The location and status of all our material is always visible throughout the
distribution network (e.g., distribution centers, transportation)
8. How have you changed your Supply chain design? [linked to the building block
containment]
a. We have not changed the structure of our supply chain since 2019.
b. We are considering sourcing material closer to our production/distribution sites.
c. We have changed towards localized sourcing for some of our material.
d. We have changed to localized sourcing for most of our material.
e. All material supply is localized to our production/distribution sites.
9. How have you changed your supplier base to increase your resilience? [linked to
the building block multisource]
a. We have not changed the structure of our supply chain since 2019.
b. We are considering increasing the number of suppliers for certain material.
c. We have tried to find new sources for some material.
d. We have added new sources for some critical material.
e. We have added new sources for the most critical material.
10. Have you changed your inventory management to increase your ability to deliver?
a. We have reduced the target level of our buffer since 2019.
b. We have not changed the way we dimension our buffers since 2019.
c. We have increased our target buffers since 2019.
d. We have updated our buffer policies since 2019 to accommodate disruptions in
a better way.
e. We use AI and machine learning to dimension our buffers.
11. To which extent have you changed your products to increase your ability to deliver?
[linked to the building block modular design]
a. We have not changed the structure of our bill of material since 2019.
b. We are considering changing the bill of material to enable more sources of same
material.
c. We have changed the bill of material for some of our materials.
d. We have changed the bill of material for most of our materials.
 Has the Pendulum Swinged Too Far from Just-in-Time 105

12. Do you have a governance model in place enabling business review with your
suppliers within Electronics?
a. We do not regularly engage with our suppliers.
b. We have a yearly review with our suppliers.
c. We have regular engagement with our suppliers on business outlook.
d. We have regular engagement with our suppliers on business outlook and new
products.
e. We have regular engagement with our suppliers on Executive level.

4 Results
All the companies in this study reported that they had their ability to meet customer
demand on-time affected during the period from 2019–2022, see Table 4. To meet cus-
tomer demand at normal lead times, supply chain changes were needed. Company C,
operating with long project lead-time, already had long lead-times which acted as a
buffer in this case. Other companies were affected so they had to prolong their lead-
times to the customers and Company D also deemed the situation as their capability to
deliver had been disturbed.

Table 4. To what extent has your ability to deliver been affected by the general shortage of
electronics?

To what extent has your ability to deliver been affected by the general A B C D E
shortage of electronics?
We have been able to supply all demand on normal lead times with no
changes in our supply chain
We have been able to supply all demand on normal lead times with x x x
changes in our supply chain
We have been able to supply all demand with prolonged lead time x x
Our capability to deliver has been disturbed x

According to Table 5, Supply chain design was not changed by large, these changes
normally take more than a year to execute. However, Company B already has a regional
supply chain. Company C within the defence industry has geopolitical regulations to
comply with, such as country of origin etc. The changes that some companies made were
more related to supply chain tactics. Company C insourced its logistics of electronics
to manage the increased buffer on some components and to increase the supply chain
visibility. Company E answered ‘other’ and clarified that thet had changed their supply
chain planning from make-to-order to make-to-stock.
In combination, the majority of the companies in this study did significant increase
in their buffer policies to cope with higher uncertainty, see Table 6. This is also supported
by the interview with the semiconductor distributor, who saw increases all across the
industries they serve in Europe. They saw a clear move from working with just-in-time
to having an inventory just-in-case.
106 J. Bäckstrand and A. Malmstedt

Table 5. How have you changed your Supply chain design?

How have you changed your Supply chain design? A B C D E

We have not changed the structure of our supply chain since 2019
We are considering to source material closer to our production/distribution x x x
sites
We have changed towards localized sourcing for some of our material
We have changed to localized sourcing for most of our material x
All material supply is localized to our production/distribution sites
Other x

Table 6. Have you changed your inventory management to increase your ability to deliver?

Have you changed your inventory management to increase your ability to A B C D E

deliver?
We have reduced the target level of our buffer since 2019 x
We have not changed the way we dimension our buffers since 2019
We have increased our target buffers since 2019 x x x
We have updated our buffer policies since 2019 to accommodate disruptions x x x
in a better way
We use AI and machine learning to dimension our buffers

Another factor which takes time to change is the supplier base. From the companies
in the study, a scattered view was presented with a trend towards an increase in the
number of suppliers of critical material, where it was possible. The progress is higher
where the spend of electronics was higher and the majority had added suppliers to critical
material, see Table 7.

Table 7. Have you changed your supplier base to increase your resilience?

Have you changed your supplier base to increase your resilience? A B C D E

We have not changed the structure of our supply chain since 2019 x
We are considering increasing the number of suppliers for certain material x x
We have tried to find new sources for some material x x
We have added new sources for some critical material x x x
We have added new sources for most critical material x x

Related to adding new sources, it often comes with changes in the Bill of Material
to accommodate other component characteristics such as size, thermal and volage etc.
 Has the Pendulum Swinged Too Far from Just-in-Time 107

This change comes with a bigger effort since it requires R&D resources. Only one
company reported that the BOMs with critical material were changed to accommodate
more components. Other saw it as one tool in the toolbox which they had tried and
considering but resource heavy, see Table 8.

Table 8. To which extent have you changed your products to increase your ability to deliver?

To which extent have you changed your products to increase your ability to A B C D E
deliver?
We have not changed the structure of our bill of material since 2019 x x
We are considering changing the bill of material to enable more sources of x
same material
We have changed the bill of material for some of our material x x
We have changed the bill of material for most of our material x

On supplier governance, Table 9, all companies had regular engagements with their
supplier. Common agenda points were business outlook and new product development.
Only one company reported that they addition to this add a governance on executive
level with their semiconductor suppliers.

Table 9. Do you have a governance model in place enabling business review with your suppliers
within Electronics?

Do you have a governance model in place enabling business review with your A B C D E
suppliers within Electronics?
We do not regularly engage with our suppliers x
We have yearly review with our suppliers x
We have regular engagement with our suppliers on business outlook x x x x
We have regular engagement with our suppliers on business outlook and new x x x x x
products
We have regular engagement with our suppliers on Executive level x x x x

When it comes to supply chain control, Table 10, all companies had knowledge of
the origin source of the material sourced. The majority of the companies had knowledge
of where the supplier base was located. When it comes to supply chain data such as
actual inventory, lead-times, delivery dates etc., only the two larger companies had that
knowledge, who also sourced directly from the first-tier suppliers which contributed to
that supply chain visibility. The actual status of a material through the entire supply
network only one of the large companies had that knowledge.
The companies that have a large part of their spend in semiconductors also reported
a relatively lower increase in material prices during this period. This stands out since the
108 J. Bäckstrand and A. Malmstedt

Table 10. What is your knowledge about other actors within semiconductors in your supply chain
and availability of data?

This section concerns supply chain control tower and your knowledge about A B C D E
other actors within semiconductors in your supply chain and availability of
data?
We know the original source from where our material is sourced from x x x x x
We are well aware where our suppliers are located x x x x
We always have information on inventory availability, lead times and x x
delivery dates in our material supply
The location and status of some of our material is always visible throughout x
the distribution network (e.g., distribution centers, transportation)
The location and status of all our material is always visible throughout the
distribution network (e.g., distribution centers, transportation)

overall price of these categories increased significantly during this period. One expla-
nation factor is that the majority of the material was sourced from the first tier which
enables another business dialog.
From the interview with the service provider of semiconductors, Company Z, we
learned that the majority of the volume of semiconductors is supplied via distributors.
Although 20% of the volume value is via first tier supplier supply for industries in handset,
telecom, and cloud. One famous example is Texas Instruments, a manufacturer of analog
chips who only engage with 10 customers directly of ca 10 000 of their customers. Hence
many companies lack direct engagement with their first tier which negatively impacts
their leverage when it comes to shortage situations and supply chain visibility.
The prices of the broad semiconductors had an annual price increase of 10–15%
were the driving forces in the beginning of the period was high demand and later on
energy prices, interest, raw material price increase and labor cost increases. On top of
this many companies had to pay an expedite fee to get priority of access material via
the spot market when they could not access the material via their normal channels. For
some materials the spot prices could be 100 times the normal price.
Changes in the Bill of Material (BOM) were frequent in the beginning of the period,
mainly with last-time-buy components. Changes to the BOMs to accommodate a lack of
other material were not widely seen, mainly explained by the huge design and engineer-
ing resources it would require. Although, a more cautions design choice of components
was seen from many customers (including supply capability).
Many companies increased their inventory to safeguard the supply chains during this
period. The actions were initiated during 2020 and 2021, but it was first in 2022 when the
supply increased and enabled increased buffers. Moreover, there have been several other
measures executed to increase supply and/or priority from the semiconductor suppliers.
• Increased forecasts (30%)
• BOM changes
• Increased buffers (1 month- >6 months of sales)
• Firm commitments. Long Term Agreements (12–24 months)
 Has the Pendulum Swinged Too Far from Just-in-Time 109

• Introduction of new sources

Due to the high book-to-bill, the semiconductor companies signaled sold-out capac-
ities and long lead-times which resulted in that new order had to be placed with long
lead-times for the consuming company with a peak of 90 weeks. Consequently, a huge
pull in demand was observed from the interviewee as “3 years of demand in 1 year”.
In parallel with this development, the demand for new supply chain models increased
when many companies realized how difficult it was to get a sufficient supply and
capability outlook, both short-term and long-term.

5 Analysis and Discussion

Increasing inventory, a known tool for handling uncertainty in the toolbox for supply
chain managers, has been accepted from company management to safeguard the sup-
ply chains and companies’ ability to meet customer commitments. This has also been
observed from the distribution side of the supply chain of semiconductors since the
increases in inventory have been significant, compared with earlier crises.
Supply chain control, however, stands out as a clear improvement area which could
have helped to increase supply chain visibility and provide more facts for decision making
when it comes to capability of semiconductors. Investment in a digital solution such as
a supply chain control tower or in the service of a logistics partner are recommended.
Supply chain design and changes in the supplier base are structural changes which take
time. The interviewed case companies showed progress in these areas with a clear trend
towards increased degree of multisource. It is, however, difficult to execute such changes
in the middle of a crisis.
To answer what changes companies have implemented to achieve more resilient
supply chains for semiconductors, we can conclude that the Supply chain managers
asked and received commitments from company management to increase the safety
stocks in the supply chain. There is also progress in structural changes in the supply
chain such as supply chain design and multisource. However, changing products or
product design is less executed or used with caution.
In addition to the insight that sufficient buffer levels are needed, many companies
have also realized that semiconductors are not just a component. The supply chain is
complex and requires additional processes and tools vs normal components to secure
capability.
To recap, several tools were utilized to protect and to increase the supply chain
resilience during the period of 2019–2022. One observation was that several parameters
were changed at the same time and since some have a lagging effect in time, this led to
the full effect first emerging during 2021Q4-2022Q1, when the underlying demand also
started to decline. One reflection on industry level has concluded that the revenue grew
by approximately 10% while the inventory increased 300%.
To summarize, the move from Just-in-time to Just-in-case has been evident in all the
case companies, and on an aggregated level this has led to a significant bullwhip effect
in relation to the actual change in which effect was amplified by the upstream position
of semiconductors in the supply chain, many involved actors in the supply chain and its
long lead time.
110 J. Bäckstrand and A. Malmstedt

To conclude, in the case of the semiconductor supply chain, the pendulum did in fact
swing too extreme from just-in-time to just-in-case.

References
1. Schulh C., Schnellbächer, W., Triplat, A., Weise, D.: The semiconductor crisis should change
your long term supply chain strategy. Harward Bus. Rev. (2022)
2. Leprince-Ringuet, D.: The global chip shortage is a much bigger problem than everyone
realised. And it will go on for longer, too (2021)
3. Miller, C.: Chip War: The Fight for the World’s Most Critical Technology. Simon and Schuster
(2022)
4. Mohammed, A., Khan, S.A.: Global disruption of semiconductor supply chains during
COVID-19: an evaluation of leading causal factors. In: 17th International Manufacturing
Science and Engineering Conference. American Society of Mechanical Engineers (2022)
5. Fitch, A.: For chip makers, the flip from shortage to glut intensifies. Wall Street J. (2022)
6. Sohn, J.: Samsung forecasts lowest earnings since 2009. Wall Street J. (2023)
7. Fitch, A.: Intel slumps on disappointing earnings amid PC weakness. Wall Street J. (2023)
8. Sohn, J.: Memory-chip makers say recovery is in sight after long slump. Wall Street J. (2023)
9. Sheffi, Y.: Building a Resilient Supply Chain in Harvard Business Review - Supply Chain
Strategy, pp. 1–4. Harvard Business School Publishing (2005)
10. Malmstedt, A., Bäckstrand, J.: How to predict disruptions in the inbound supply chain in a
volatile environment. In: 10th Swedish Production Symposium, Skövde, Sweden (2022)
11. Christopher, M., Peck, H.: Building the resilient supply chain (2004)
12. Chopra, S., Sodhi, M.: Supply-chain breakdown. MIT Sloan Manag. Rev. 46(1), 53–61 (2004)
13. Lund, S., et al.: Risk, resilience, and rebalancing in global value chains. McKinsey Global
Institute (2020)
14. Shih, W.: Is it time to rethink globalized supply chains? MIT Sloan Manag. Rev. 61(4), 1–3
(2020)
15. Goering, K., Kelly, R., Mellors, N.: The next horizon for industrial manufacturing: adopting
disruptive digital technologies in making and delivering. In: Digital McKinsey, pp. 1–10
(2018)
16. Ivanov, D., Dolgui, A.: A digital supply chain twin for managing the disruption risks and
resilience in the era of Industry 4.0. Prod. Plan. Control, 1–14 (2020)
17. Linton, T., Vakil, B.: Coronavirus is proving we need more resilient supply chains. Harvard
Bus. Rev. 5, 121–129 (2020)
18. Lee, H.L., Padmanabhan, V., Whang, S.: The bullwhip effect in supply chains. Sloan Manag.
Review 38(3), 93–102 (1997)
19. Kraljic, P.: Purchasing must become supply management. Harv. Bus. Rev. 61(5), 109–117
(1983)
20. Ismail, A., Ehm, H.: Simulating and evaluating the bullwhip effect along an end-to-end semi-
conductor automotive supply chain amid COVID-19 crisis. In: Anylogic Conference 2021
(2021)
21. Gupta, M.C., Boyd, L.H.: Theory of constraints: a theory for operations management. Int. J.
Oper. Prod. Manag. 28(10), 991–1012 (2008)
22. TechTarget. Semiconductor definition. WhatIs? (2021). https://www.techtarget.com/whatis/
definition/semiconductor
23. Screen. Semiconductor manufacturing processes. https://www.screen.co.jp/spe/en/process
24. Matsusada Precision. The semiconductor manufacturing process (front-end process).
TechTips (2022). https://www.matsusada.com/column/sc_mfg_proces-front-end.html
 Has the Pendulum Swinged Too Far from Just-in-Time 111

25. Heaven, D.: The humble mineral that transformed the world. Made on Earth (2021) https://
www.bbc.com/future/bespoke/made-on-earth/how-the-chip-changed-everything/
26. Moore. Moore’s law (2023). https://en.wikipedia.org/wiki/Moore%27s_law
27. Weiwen, H.: Of Chips and Trade, in ChinaUS Focus (2022). https://www.chinausfocus.com/
finance-economy/of-chips-and-trade
28. Frieske, B., Stieler, S.: The “Semiconductor Crisis” as a result of the COVID-19 pandemic
and impacts on the automotive industry and its supply chains. World Electr. Veh. J. 13(10),
189 (2022)
 Integrating Lean, Agile, Resilient and Green
Supply Chain Management
in Engineer-to-Order Contexts: Insights
from Expert Interviews

Antonio Masi(B) and Margherita Pero

Politecnico di Milano, Milano, Italy

[email protected]

Abstract. “LARG” is a supply chain management paradigm that combines dif-

ferent managerial approaches to achieve waste reduction (Lean), rapid response
to customer requests (Agile), resilience to disruptions (Resilient), and environ-
mental sustainability (Green). LARG practices could help addressing some of
the strategic priorities of Engineer-to-Order (ETO) contexts, such as the need for
responding quickly to customer requests, while assuring resiliency in a volatile
context like the current one. However, its feasibility in ETO contexts is still under-
studied. The purpose of this paper is to open a debate on this topic. To do that, we
develop a conceptual framework of the main LARG practices, based on a litera-
ture review, and then we validate it through four expert interviews, with managers
of Italian machinery companies. Our results suggest three main managerial rec-
ommendations. First, they show the benefits of servitization for both resilience
and sustainability, thanks to services, like remote monitoring, that allow to reduce
material and energy wastes, while limiting the need for travelling. Second, they
stress the role of modularity as an antecedent both of “leagile” and of servitization
in ETO contexts. Third, they call for a higher adoption of green practices, which
are much less widespread than lean, agile, and resilient ones, despite some early
signals of change. Thus, this paper addresses the call for further research on sup-
ply chain paradigms implementation in ETO contexts, and opens new avenues for
future research.

Keywords: Engineer-to-Order · Supply Chain · Sustainability · Resilience ·

LARG

1 Introduction
Nowadays’ supply chains face the challenge of being resilient and environmentally
sustainable, while assuring the right product, at the right time, to the right customer,
at the right cost. Disruptive events, such as the Covid-19 pandemic and the Russo-
Ukrainian war, caused a shortage of raw materials and components, that put a strain on
the resilience of global supply chains. Meanwhile, climate change calls for a reduction
© IFIP International Federation for Information Processing 2023
Published by Springer Nature Switzerland AG 2023
E. Alfnes et al. (Eds.): APMS 2023, IFIP AICT 691, pp. 112–125, 2023.
https://doi.org/10.1007/978-3-031-43670-3_8
 Integrating Lean, Agile, Resilient and Green Supply Chain Management 113

of greenhouse gases emissions and resource depletion, thus requiring supply chains to
become “greener”.
Engineer-to-Order (ETO) supply chains are affected by disruptions too, while also
being called to take environmental concerns into their agenda. ETO contexts are char-
acterized by the penetration of customer orders into production, design and engineering
processes: depending on the extent of order penetration in the design, engineering and
production, ETO companies can be classified in different engineering and production
decoupling configurations [1]. In line with Cannas et al. [1], in this study we consider as
ETO companies also those for which only a fraction of the design and engineering pro-
cess is performed based on order, e.g. Configure-to-Order companies. ETO supply chains
deliver complex, capital intensive and technologically innovative products. Typical ETO
sectors are, among the others, construction, machinery, and shipbuilding. Construction
strongly contributes to the emission of greenhouse gas in the atmosphere [2]. Machinery
and shipbuilding were strongly affected by the shortages of materials and restrictions to
work activities [3, 4].
To face such emerging challenges, companies need to find a way to implement
practices that allow them to be, at the same time, lean and agile in the response to
customer orders, while pursuing resiliency and sustainability. In line with Naylor [5],
the Lean paradigm is about eliminating waste, while the Agile paradigm is about being
responsive to customer requests [6]. There have been attempts over the years to study
how to combine different supply chain paradigms. Supply chain paradigms are sets of
practices aligned to a cultural and philosophical belief [7, 8]. In the 90s, the “Leagile”
paradigm was introduced, as a combination of Lean and Agile [9]. Later, researchers
investigated how to combine Lean or Agile and Green paradigms [6]. Such combination
requires companies to be able to manage trade-offs. In fact, some practices belonging to
different paradigms are supporting each other, while other ones are in trade-off [6].
Until 2011, the different supply chain paradigms had been studied in isolation, or
as combination of two. In 2011, Carvalho et al. [10] suggested that companies need to
pursue a paradigm that combines all the ones presented by previous researchers: Lean,
Agile, Resilient and Green. Therefore, they conceptualized the LARG paradigm, as the
integration of the Lean, Agile, Resilient and Green paradigms, in which synergies and
divergences are present.
The LARG paradigm may help address nowadays’ supply chains challenges, includ-
ing the ones faced by ETO supply chains. LARG practices can lead to contradictory per-
formance outcomes, thus making important to take into account the peculiarities of each
supply chain to understand which practices could be actually beneficial and which ones
not [12, 13]. Recent studies show the feasibility of using the LARG framework in the
agri-food industry: Bottani et al. [14] propose a performance measurement framework
for the adoption of LARG in agri-food supply chains, while Sahu et al. [15] discuss
the barriers to the adoption of LARG in such contexts. The impact of the use of LARG
practices on performance depends also by the use of digital tools, e.g. data analytics [15,
16].
However, although the literature on LARG has grown in the past decade, the deploy-
ment of this paradigm still needs to be clarified in ETO contexts. In ETO supply chains,
the implementation of Lean has been investigated in isolation [16], or in combination
114 A. Masi and M. Pero

with Agile (Leagile) [17], and Green. For instance, Ramirez-Peña et al. [18], conclude
that, under the Industry 4.0 perspective, shipbuilding companies should opt for both
Lean and Green. Furthermore, Centobelli et al. [19], show how Industry 4.0 supports
higher supply chain resilience in ETO contexts. However, few papers discuss whether
the integration of all the four paradigms (LARG) could be feasible and/or beneficial in
ETO supply chains, and little is known especially for the machinery industry.
Given the recent importance of pursuing contrasting objectives, i.e. being resilient
and green, while reducing waste and being able to quickly respond to customer needs,
with this paper we explore whether and how machinery companies deploy in supply
chain related practices, different strategies, i.e. Lean, Agile, Resilient and Green. Hence,
we address the following research question (RQ): “Do ETO machinery companies adopt
a combination of lean, agile, resilient, and green supply chain practices? And how?”.
To seek an answer to our RQ, we develop a theoretical framework of LARG supply
chain practices, based on a literature review, and then we try to validate it and improve
it through expert interviews in the Italian machinery industry. We chose to focus on this
industry, because it faced huge disruptions during the pandemic, since Italy was one of
countries first and most affected by COVID-19 [20].
The aim of this paper is to gather preliminary insights from literature and experts, to
open a discussion and further research on this topic. Hence, our contribution is to explore
the feasibility of the LARG supply chain practices in ETO. In this, we also address the
recent call for further research on lean and agile practices in ETO [21].
The remainder of the paper is structured as follows: Sect. 2 presents the theoretical
background and the research framework, including the LARG supply chain practices
identified though a literature review. Section 3 describes the methodology used to validate
and refine the framework, namely expert interviews. Results are presented in Sect. 4 and
discussed in Sect. 5. Finally, Sect. 6 shows the conclusions and the directions for future
research.

2 Theoretical Background

2.1 Overview of the LARG Paradigm

“LARG” can be defined as a supply chain “paradigm” [11], pursuing different SC

objectives (i.e. to be lean, agile, resilient, and green) by adopting various supply chain
management practices [22].
Azevedo et al. [22] list 11 LARG practices, including just-in-time practices (lean),
fast responsiveness to changing market needs (agile), lead time reduction (resilient), and
reduction of materials employed in product manufacturing (green).
Carvalho et al. [11] analyze the interdependencies between 27 LARG practices, by
using causal loop diagrams. They find that LARG practices have both synergies and
trade-offs. All the LARG practices increase information exchanges frequency and sup-
ply chain integration level, while decreasing production and transportation lead times.
However, they have divergent impacts on capacity surplus, inventory level, and replen-
ishment frequency. For instance, lean and green practices reduce capacity surplus to
limit wastes and improve the environmental footprint, while the agile and resilient ones
 Integrating Lean, Agile, Resilient and Green Supply Chain Management 115

increase it, in order to satisfy more quickly customer needs, and be better prepared to
face SC disruptions.
Hence, it is necessary to carefully choose which LARG practices to implement,
depending on the competitive priorities of each company. For this reason, Cabral et al.
[12] propose an analytic network process model to choose the most appropriate LARG
practices and the related Key Performance Indicators (KPIs). However, their study takes
into account only 3 LARG practices, and is limited to the automotive sector.
Building upon the KPIs discussed by Cabral et al. [12], Azevedo et al. [13] propose
a “LARG” index. Such index can be used to monitor the leanness, agility, resilience
and greenness of single companies or of whole supply chains, like the automotive one.
Their results further validate the concept that companies usually adopt different LARG
practices to varying extents.

2.2 LARG in ETO Supply Chains

To the best of our knowledge, few studies have discussed the adoption of LARG practices
in ETO supply chains. The main exception is the work by Ramirez-Peña et al. [18],
who analyze how Industry 4.0 technologies support leanness, agility, resilience and
greenness in a typical ETO sector: shipbuilding. Their results reveal that the use of
lean and green practices can result in performance improvement. However, they take
into little consideration the fact that the shipbuilding supply chain adopts an ETO order
fulfillment strategy.
We argue that LARG practices could be interesting for ETO supply chain manage-
ment, because they can address some of the key issues of such supply chains. In fact, all
the LARG practices support information exchanges and supply chain integration [11],
which are also strategic priorities for ETO supply chains [10, 19]. Supply chain inte-
gration is critical to properly manage the interdependencies between activities, which
are frequent in complex projects [24]. Furthermore, information management is key
to improving production planning and control [25] and risk management [26]. More-
over, all the LARG practices decrease production lead times [11], which is especially
important for ETO companies.
However, that the impact of LARG practices on inventory levels, capacity surplus,
and replenishment frequency is hard to establish a-priori. Yet, the strategic management
of capacity is of particular interest for ETO contexts, being a key managerial lever to plan
the production upstream of the CODP [27]. Hence, an empirical study on the feasibility
of LARG practices in ETO contexts is necessary, and this is the gap which the present
paper aims at addressing.

2.3 Research Framework

To understand whether and how LARG practices are feasible in ETO supply chains, we
develop a preliminary research framework. The framework summarizes the main LARG
practices discussed in the literature reviewed in Sect. 2.1, and is the basis for the expert
interviews. The interviews aim at testing the adoption of extant LARG practices in ETO
SCs, and at identifying new ones.
116 A. Masi and M. Pero

The research framework is schematized in Table 1. The first column summarizes the
main LARG practices emerged from the literature review. The central columns show
whether each practice supports each objective of the LARG paradigm (+), or not (−). The
“ =” symbol means that a practice is neither supportive nor detrimental to an objective.
It can be noted that the adoption of some practices might be supportive of one objective
while being detrimental of others, thus generating trad-offs to manage.
With respect to extant literature, our framework has two main differences. The first
one is that we eliminated, or modified, the practices that could be confused with KPIs. For
instance, we do not refer to “cycle/setup time reduction” [22] but rather to “cycle/setup
time reduction practices” like the Single Minute Exchange Die (SMED) [28]. We made
this change in order to make our interviews more focused on what companies do, rather
than on their objectives. The second one is that we often clustered different practices
into a single “bundle”. For instance, we refer to “supply chain collaboration”, bearing
in mind that it refers to different practices (e.g., increasing trust, information sharing,
profit sharing), listed by Azevedo et al. [13]. We made this change to let our discussion
with interviewees be more open, thus possibly revealing new practices.

Table 1. Preliminary list of LARG practices

Practices L A R G References
Just in time production + + = = [11–14, 28]
Supply chain collaboration practices + = = + [11–14, 28]
Cycle/setup time reduction practices + + = = [11–14, 28]
Stocks minimization + − − + [11–13]
Surplus production capacity − + + − [12, 28]
Sourcing strategies to allow switching suppliers − + + = [13]
Reconfigurable production processes = + + = [11–13]
Eco-design of products + = = + [11, 13, 28]
Transportation lead times reduction practices + = = + [12, 14]
Material and energy waste reduction practices + = = + [11–14, 28]

Several noteworthy observations can be made based on the table. First, some prac-
tices support multiple LARG objectives without negatively affecting others, while other
practices have conflicting impacts. For instance, “Just in time production” and “Cy-
cle/setup time reduction practices” support Lean and Agile without significantly damag-
ing Resilient and Green, while "Stocks minimization” favors Lean and Green objectives,
but has a negative impact on Agile and Resilient objectives.
Second, certain couples of objectives are usually more synergistic than others. For
example, it is worth noting how all the practices that support Green objectives also
support Lean ones, and vice versa, as the case of “Surplus production capacity” shows.
The reason for this is that higher efficiency often leads also to lower resource consumption
and, thus, higher sustainability. Similarly, positive impacts on Resilience come with
 Integrating Lean, Agile, Resilient and Green Supply Chain Management 117

benefits for Agility as well, in 3 cases out of 5. In fact, higher operational responsiveness
can be used to react both to customer requests and to unexpected disruptions.
Last but not least, it is interesting to note that few practices found in the literature
about LARG deal with engineering, which is key for ETO companies. The main excep-
tion is the eco-design of products, which supports Lean and Green objectives. Studies in
the shipbuilding industry show that technologies like simulation can support the design
of more sustainable products by reducing energy wastes [29]. It is interesting to study
how this practice is adopted in ETO machinery firms, too.

3 Methodology
We explore whether and how ETO companies implement the LARG paradigm through
expert interviews. To capture the perspective of the Italian machinery industry, we
interviewed representatives of four ETO companies located in Northern Italy.

3.1 Selection of the Experts

We selected the experts based on convenient sampling. Given the declared intention
of focusing on the machinery industry in Northern Italy, we extracted a list of tool-
ing machines manufacturers from the AIDA database (https://www.bvdinfo.com/it-it/
le-nostre-soluzioni/dati/nazionali/aida), similarly to what other papers about machinery
ETO companies did [1, 27]. Most of the shortlisted companies were Small-to-Medium
Enterprises (SMEs). Out of this list, as a first action, we contacted those active in the
areas of Milano and Monza. The choice of these areas was due to their high industrial-
ization, associated with the fact that the area was strongly hit by COVID-19 pandemic
and associated restrictions. Then, we sent an email to these companies inviting them to
participate to the research. In total, we contacted 22 companies. Four of them accepted
to identify an expert for the interview. Most experts have more than 10 years experience
in the field, as shown in Table 2. The experts had different roles in the companies, to
collect different perspectives.

3.2 Data Gathering and Analysis

To conduct the interviews, we defined a semi-structured interview guideline. We opted

for open-ended questions to allow new concepts to emerge. We organized the interview
guideline in the following sections:
1. general overview of the expert’s experience and role;
2. practices used in ETO companies to eliminate waste;
3. practices used in ETO companies to increase responsiveness;
4. practices used in ETO companies to achieve higher resilience;
5. practices used in ETO companies to improve the environmental footprint.
Each interview lasted about one hour. Due to logistics issues, all of them were con-
ducted online, on the Microsoft Teams platform. Then, they were recorded, transcribed,
118 A. Masi and M. Pero

Table 2. Overview of the sampled experts and of their companies

Expert Role in the Years of experience Main product of the Turnover of the
company company company (Me, in
2022)
Expert 1 ICT director About 20 Sheet metal 45
working plants
Expert 2 Purchasing N.A. Shot blasting 18
manager machines
Expert 3 Innovation About 10 Automated 6
manager industrial plants
Expert 4 Chief Operating About 40 Automating 12
Officer dispensing systems

and then saved in a folder shared among the authors. For what concerns the data analy-
sis, we started with a deductive coding process, to identify the main practices for each
category of the LARG paradigm mentioned in the interview. Later, we also adopted an
inductive coding approach, to let new aspects emerge from data.

4 Results
Table 3 provides an overview of the LARG practices emerged from the expert inter-
views. Most of them were identified in the preliminary framework. Remarkably, two
practices discussed in the literature (“surplus production capacity” and “collaborative
eco-design of products”) did not emerge from our interviews. However, other practices,
less discussed in the literature, emerged. In the following paragraphs, we will comment
the practices, dividing them on the basis of the paradigm to which they mainly refer to.

Table 3. Overview of the LARG practices discussed by the experts interviewed

LARG practices Experts

Preliminary list 1 2 3 4
Just in time production X
Supply chain collaboration practices X X X X
Cycle/setup time reduction practices X X
Safety stocks minimization X X X
Surplus production capacity
Sourcing strategies to allow switching suppliers X X X
Reconfigurable production processes X X X
(continued)
 Integrating Lean, Agile, Resilient and Green Supply Chain Management 119

Table 3. (continued)

LARG practices Experts

Preliminary list 1 2 3 4
Collaborative eco-design of products
Transportation lead times reduction practices X
Material and energy waste reduction practices X X X
Other practices 1 2 3 4
Improved planning systems X X X X
Aggregated purchase orders X X X X
Purchase of sub-assemblies X X
Servitization X X X

4.1 Lean Practices

The main lean practices emerged from the interviews regard inventory management:
safety stocks minimization, and aggregated purchase orders.
In general, experts agreed on the need to minimize safety stocks: “we stock only
low-value items, while we purchase the high-value ones time after time” (Expert 1);
“we keep very few stocks, but we have a large warehouse of spare parts” (Expert 2);
“for the standard groups, we minimize stocks, and replenish them with an economic
order quantity approach” (Expert 3).
However, interviews also showed that a certain amount of stocks is still necessary.
In this sense, experts pointed out some approaches to reduce costs as much as possible.
For instance, they highlighted the importance of trying to aggregating purchase orders,
to leverage scale economies. Modularity and standardization are synergistic with such
approach: “if I standardize the modules, the impact on the supply chain is that I do
not need to ask for different products every single time… Instead, I can ask for larger
batches, stock them, and maybe use them for spare parts” (Expert 3). However, as
Expert 1 points out, “aggregating volumes is not easy, because it clashes against our
just-in-time approach”. Vendor Managed Inventory (VMI) and similar collaborative
practices can mitigate this issue: “at the beginning of the year, I will tell my supplier I
will purchase 1000 pieces, at a pre-determined price. Then, I will receive these pieces in
smaller batches, scheduled on different dates, while benefitting from the initial discount”
(Expert 2).
Practices like VMI require supply chain collaboration. Concerning this, all the inter-
viewed companies established long-term (10+ years), trustworthy relationships with key
suppliers. Usually, their partners are SMEs, to which they outsource the production of
customized mechanical components or sub-assemblies, “products that require a know-
how that the supplier needs to develop over time, and you cannot develop it in one week,
but rather in one year” (Expert 1).
It is worth remarking, though, that the small size of the suppliers can be a double-
edged sword for collaboration. On the one hand, small suppliers are “flexible”, i.e., “in
120 A. Masi and M. Pero

case of emergency, they support us, and are open to renegotiate agreements” (Expert
4). This is usually due to their low bargaining power: “some suppliers sell 80% of their
products to us. They are not our subsidiaries, but little is missing” (Expert 3). On the
other hand, they have very unstructured management processes, which may prevent joint
planning practices. In the words of Expert 1: “in some small workshops, the company
owner is also in charge of turning the lathe on. These people can be a bit naïve, not
used to structured processes and, unfortunately, prone to intuitive decision-making. If
you talk about ‘planning’, they may not even know what it is”.

4.2 Agile Practices

When it comes to the agile paradigm, all the experts agreed on the importance of a highly
customer-centric approach. This is intuitive, in ETO contexts. Interestingly though, inter-
viewees pointed out how delivery speed is more important than product customization,
for some product families.
Expert 4, for instance, describes how the company cut assembly times by parallelizing
the production of standard parts: “today, these components are stocked, and produced
during ‘dead times’. This means that they are made in moments totally unrelated to
production needs”. Clearly, this practice becomes more beneficial, the more the company
relies on product modularity and standardization. In fact, Expert 4 adds: “we make mass
customized products, rather than standard products, or purely ETO ones”.
Another critical aspect is whether the assembly processes are automated or manual,
the latter ones being the most agile solutions: “our assembly processes are extremely
flexible because they are totally manual… and they can be reconfigured in little time,
even a few hours” (Expert 3).

4.3 Resilient Practices

A key resilient practice emerged from the interviews concerns sourcing. Companies
with few backup suppliers faced a lot of challenges during the pandemic: “the pur-
chasing department suffered because, while two years ago it was possible to engage
different suppliers, now this possibility is very limited, because they are overwhelmed
by requests… So, suppliers can blow hot and cold” (Expert 1). Conversely, companies
with more backup suppliers had advantages in terms of resilience, as well as of lower
costs, thanks to the higher competition between them.
An interesting trend related to the pandemic was servitization. Servitization is the
transformation process manufacturers undergo when moving from providing products
to providing increasingly advanced services [30]. We noticed that companies offering
services on top of their products managed to keep part of their operations up and running,
regardless of the limitations caused by the pandemic. For instance, Expert 3 discussed the
advantages of remote monitoring: “during the pandemic, we managed to test machines
remotely: the machine was physically here, while the client could watch the test thanks
to the remote monitoring system. Vice versa, some machines were shipped to clients, and
we supported them, remotely”.
 Integrating Lean, Agile, Resilient and Green Supply Chain Management 121

4.4 Green Practices

The majority of the experts agreed on a limited adoption of green practices. For instance,
Expert 1 said that they “just respect the law, but do not go beyond it”, while Expert 2
added that their company aimed at receiving the ISO 14001 certification for energy
savings.
As a matter of fact, most of the initiatives leading to higher sustainability were
undertaken only because they were synergistic with other paradigms, like the lean one.
For example, different experts made attempts to reduce material wastes, which have
become critical after the pandemic: “in our case, the price of raw materials has doubled”
(Expert 1). For instance, Expert 4’s company started to recycle the off-cuts of tubes to
create new ones, and minimize the purchases of cables, or other “bulk” items.
More impactful actions have been discussed by Expert 3 and, remarkably, they are
interlinked with the servitization trend. For example, remote monitoring allows a fast
identification of material and energy wastes in the assembly lines produced by Expert 3’s
company. In addition, it lets the company help the client’s maintenance staff remotely,
thus cutting transportation costs, and improving the environmental footprint. Moreover,
huge savings of materials are possible thanks to “revamping” services: “if the machine
is designed according to the principles of modularity, tomorrow you will not need to buy
a new machine, but you will just bear the costs of its reconfiguration… This happens in
a fast way, with limited costs and, above all, material savings, because I do not need to
make a brand new production station, but I just have to customize it” (Expert 3). These
words suggest a link between sustainability, servitization, and product modularity.

5 Discussion

In this section, we compare our results with previous studies, and summarize our
contributions in three practices that could be particularly beneficial for ETO firms in
implementing LARG paradigm.

5.1 Servitize Your Business Model

An unexpected result from our interviews was the rise of servitization in ETO contexts
and its positive impact on resilience and sustainability. For instance, experts discussed
how services like remote monitoring allowed to reduce material and energy wastes,
and kept operations up and running despite the disruptions caused by the COVID-19
pandemic.
In terms of resilience, our results are aligned with the survey by Rapaccini et al.
[20], in which 57% of the respondents claimed that their innovation initiatives related to
advanced services were highly accelerated by the pandemic. However, the same study
points out that basic services (e.g., reactive maintenance, repair, and training) were more
affected by the pandemic. This is in line with previous studies, suggesting that industrial
services are vulnerable to uncertainty [31].
In terms of sustainability, our results support mainstream literature, according to
which servitization has a positive environmental impact [32]. However, more recent
122 A. Masi and M. Pero

studies warn against such generalizations. For instance, Hao et al. [33] conclude that
the impact of both lean and servitization practices on sustainability is much more com-
plicated than expected, while Abdelkafi et al. [34] highlight the importance of having a
system-wide perspective when assessing the effect of servitization on sustainability.

5.2 Modularize Your Products

Based on our results and on extant literature, we found a positive impact of product mod-
ularity on all the LARG practices. This has important implications for the engineering
process of ETO companies.
First, we noticed a synergy between modularity and lean approaches, like the aggre-
gation of purchase orders. In this sense, we agree with previous studies, arguing that
modularity can be antecedent of lean, not only in the machinery industry [16], but also
in other ETO contexts [35].
Moreover, we observed how modularity can reduce production lead times, for
instance by increasing the number of shared components that can be produced dur-
ing dead times, or in parallel with other pieces. This leads to improved speed and, thus,
agility. This is in line with what happens in the construction sector where agile scheduling
and control approaches can significantly decrease buffer sizes and lead times [36].
We did not find direct impacts of modularity on resilient or green practices and, to the
best of our knowledge, little research has been done in this direction. However, we noticed
that modularity can support the introduction of certain services, like revamping. This
confirms previous studies suggesting that modularity can support servitization in ETO
contexts [37], and servitization, in turn, can increase both resilience and sustainability
(5.1).

5.3 Improve Your Green Performance

Our results show a still limited adoption of green practices in ETO machinery companies.
In this sense, our results contrast with those by Ramirez-Peña et al. [18] in the shipbuild-
ing industry. However, the trends in the machinery sector are changing. For instance,
UCIMU – the Italian machine tool, robots, automation systems and ancillary products
manufacturers’ association – has recently published the first sustainability report of
the machine tool sector in Europe [38], highlighting the need for higher awareness on
Sustainable Development Goals in the sector.
Also, academic literature confirms the need for ETO sectors to become more sus-
tainable. Although several barriers – especially cultural – need to be overcome [39],
studies show that technology can be a useful tool. For instance, Strandhagen et al. [29]
discuss how Industry 4.0 can support sustainable operations in the shipbuilding sector.
Our findings concerning the benefits in terms of sustainability of services like predictive
maintenance confirm this trend.
 Integrating Lean, Agile, Resilient and Green Supply Chain Management 123

6 Conclusions

This paper discusses the adoption of LARG practices in ETO supply chains, by exploring
the adoption of lean, agile, resilient, and green supply chain practices in Italian machinery
companies. To this aim, we developed a research framework based on the literature, and
fine-tuned it by means of four expert interviews.
Our results suggest that Italian machinery companies are mostly implementing lean,
agile and resilient supply chain management practices, while limited attention is paid
to the green ones. Based on these results, we call for higher attention to green supply
chain performance, while suggesting that two main strategies, namely servitization and
modularity, can be key in reaching LARG objectives. As a matter of fact, servitization
has been found to positively affect resilience and green objectives, while modularity
supports reaching all LARG objectives, while also enabling servitization strategies.
Our paper contributes to the discussion on LARG paradigm, by exploring its appli-
cation in the ETO supply chains, answering the call for further research on supply chain
strategies in ETO [10].
This work presents some limitations. First of all, the results are based on a limited
number of experts interviews, mainly focused on the exploration of LARG practices
adoption. Further research can be devoted to enlarge the empirical base, by also exploring
the adoption of the newly identified LARG practices and their inter-relationships, for
building a comprehensive discourse around LARG deployment in ETO supply chains.
Moreover, from our perspective, future research should be devoted to investigate how
servitization and modularity can synergistically support ETO supply chains to become
more sustainable.

References
1. Cannas, V.G., Gosling, J., Pero, M., Rossi, T.: Engineering and production decoupling con-
figurations: an empirical study in the machinery industry. Int. J. Prod. Econ. 216, 173–189
(2019). https://doi.org/10.1016/j.ijpe.2019.04.025
2. World Green Building Council: New Report: the Building and Construction Sector Can Reach
Net Zero Carbon Emissions by 2050. London (2019)
3. Alfnes, E., Gosling, J., Naim, M., Dreyer, H.C.: Exploring systemic factors creating uncer-
tainty in complex engineer-to-order supply chains: case studies from Norwegian shipbuilding
first tier suppliers. Int. J. Prod. Econ. 240, 1–16 (2021). https://doi.org/10.1016/j.ijpe.2021.
108211
4. CDP: Industrial machinery and Covid-19: scenario, impacts and prospects (2023). https://
www.cdp.it/sitointernet/page/en/industrial_machinery_and_covid19_scenario_impacts_
and_prospects?contentId=TNK30415. Accessed 17 Apr 2023
5. Ben, N.J., Naim, M.M., Berry, D.: Leagility: integrating the lean and agile manufacturing
paradigms in the total supply chain. Int. J. Prod. Econ. 62, 107–118 (1999)
6. Lee, H.L.: Aligning supply chain strategies with product uncertainties. Calif. Manage. Rev.
44, 105–119 (2002)
7. Christopher, M., Towill, D.R.: Developing market specific supply chain strategies. Int. J.
Logist. Manage. 13, 1–14 (2002). https://doi.org/10.1108/09574090210806324
124 A. Masi and M. Pero

8. Ciccullo, F., Pero, M., Caridi, M., Gosling, J., Purvis, L.: Integrating the environmental and
social sustainability pillars into the lean and agile supply chain management paradigms: a
literature review and future research directions. J. Clean. Prod. 172, 2336–2350 (2018). https://
doi.org/10.1016/j.jclepro.2017.11.176
9. Naim, M., Gosling, J.: On leanness, agility and leagile supply chains. Int. J. Prod. Econ. 131,
342–354 (2011). https://doi.org/10.1016/j.ijpe.2010.04.045
10. Carvalho, H., Cruz-Machado, V.: Integrating lean, agile, resilience and green paradigms in
supply chain management (LARG_SCM). Supply Chain Manage. (2011). https://doi.org/10.
5772/14592
11. Carvalho, H., Duarte, S., Machado, V.C.: Lean, agile, resilient and green: divergencies and
synergies. Int. J. Lean Six Sigma 2, 151–179 (2011). https://doi.org/10.1108/204014611111
35037
12. Cabral, I., Grilo, A., Cruz-Machado, V.: A decision-making model for lean, agile, resilient
and green supply chain management. Int. J. Prod. Res. 50, 4830–4845 (2012). https://doi.org/
10.1080/00207543.2012.657970
13. Azevedo, S.G., Carvalho, H., Cruz-Machado, V.: LARG index: a benchmarking tool for
improving the leanness, agility, resilience and greenness of the automotive supply chain.
Benchmarking 23, 1472–1499 (2016). https://doi.org/10.1108/BIJ-07-2014-0072
14. Bottani, E., Bigliardi, B., Rinaldi, M.: Development and proposal of a LARG (lean,
agile, resilient, green) performance measurement system for a food supply chain. IFAC-
PapersOnLine 55, 2437–2444 (2022). https://doi.org/10.1016/j.ifacol.2022.10.074
15. Sahu, A.K., Raut, R.D., Gedam, V.V., Cheikhrouhou, N., Sahu, A.K.: Lean–agile–resilience–
green practices adoption challenges in sustainable agri-food supply chains. Bus. Strateg.
Environ., 1–20 (2022). https://doi.org/10.1002/bse.3299
16. Birkie, S.E., Trucco, P.: Understanding dynamism and complexity factors in engineer-to-order
and their influence on lean implementation strategy. Prod Plan Control 27, 345–359 (2016).
https://doi.org/10.1080/09537287.2015.1127446
17. Ponticelli, S., Mininno, V., Dulmin, R., Aloini, D.: Supply chain implications for one-off
luxury products: cases from the yacht industry. Int. J. Retail Distrib. Manag. 41, 1008–1029
(2013). https://doi.org/10.1108/IJRDM-01-2013-0002
18. Ramirez-Peña, M., Sánchez Sotano, A.J., Pérez-Fernandez, V., Abad, F.J., Batista, M.:
Achieving a sustainable shipbuilding supply chain under I4.0 perspective. J. Clean. Prod.
244, 1–11 (2020). https://doi.org/10.1016/j.jclepro.2019.118789
19. Centobelli, P., Cerchione, R., Maglietta, A., Oropallo, E.: Sailing through a digital and resilient
shipbuilding supply chain: an empirical investigation. J. Bus. Res. 158, 113686 (2023). https://
doi.org/10.1016/j.jbusres.2023.113686
20. Rapaccini, M., Saccani, N., Kowalkowski, C., Paiola, M., Adrodegari, F.: Navigating disrup-
tive crises through service-led growth: the impact of COVID-19 on Italian manufacturing
firms. Ind. Mark Manage. 88, 225–237 (2020). https://doi.org/10.1016/j.indmarman.2020.
05.017
21. Cannas, V.G., Gosling, J.: A decade of engineering-to-order (2010–2020): progress and
emerging themes. Int. J. Prod. Econ. 241, 1–19 (2021). https://doi.org/10.1016/j.ijpe.2021.
108274
22. Azevedo, S.G., Carvalho, H., Cruz-Machado, V.: A proposal of LARG supply chain manage-
ment practices and a performance measurement system. Int. J. e-Educ., e-Bus., e-Manage.,
e-Learn. 1, 7–14 (2011). https://doi.org/10.7763/ijeeee.2011.v1.2
23. Gosling, J., Naim, M.: Engineer-to-order supply chain management: a literature review and
research agenda. Int. J. Prod. Econ. 122, 741–754 (2009). https://doi.org/10.1016/j.ijpe.2009.
07.002
 Integrating Lean, Agile, Resilient and Green Supply Chain Management 125

24. Mello, M.H., Strandhagen, J.O., Alfnes, E.: The role of coordination in avoiding project delays
in an engineer-to-order supply chain. J. Manuf. Technol. Manage. 26, 429–454 (2015). https://
doi.org/10.1108/JMTM-03-2013-0021
25. Adrodegari, F., Bacchetti, A., Pinto, R., Pirola, F., Zanardini, M.: Engineer-to-order (ETO)
production planning and control: an empirical framework for machinery-building companies.
Prod. Plan. Control 26, 910–932 (2015). https://doi.org/10.1080/09537287.2014.1001808
26. Sylla, A., Coudert, T., Vareilles, E., Geneste, L., Aldanondo, M.: Possibilistic Pareto-
dominance approach to support technical bid selection under imprecision and uncertainty
in engineer-to-order bidding process. Int. J. Prod. Res. 59, 6361–6381 (2021). https://doi.org/
10.1080/00207543.2020.1812754
27. Cannas, V.G., Gosling, J., Pero, M., Rossi, T.: Determinants for order-fulfilment strategies in
engineer-to-order companies: insights from the machinery industry. Int. J. Prod. Econ. 228,
1–17 (2020). https://doi.org/10.1016/j.ijpe.2020.107743
28. Anvari, A.R.: The integration of LARG supply chain paradigms and supply chain sustainable
performance (a case study of Iran). Prod. Manuf. Res. 9, 157–177 (2021). https://doi.org/10.
1080/21693277.2021.1963349
29. Strandhagen, J.W., Buer, S.V., Semini, M., Alfnes, E., Strandhagen, J.O.: Sustainability chal-
lenges and how Industry 4.0 technologies can address them: a case study of a shipbuilding
supply chain. Prod. Plan. Control 33, 995–1010 (2022). https://doi.org/10.1080/09537287.
2020.1837940
30. Baines, T., Bigdeli, A.Z., Sousa, R., Schroeder, A.: Framing the servitization transformation
process: a model to understand and facilitate the servitization journey. Int. J. Prod. Econ. 221,
1–16 (2020). https://doi.org/10.1016/j.ijpe.2019.07.036
31. Durugbo, C., Erkoyuncu, J.A.: Mitigating uncertainty for industrial service operations: a multi
case study (2016)
32. Blüher, T., Riedelsheimer, T., Gogineni, S., Klemichen, A., Stark, R.: Systematic literature
review-effects of PSS on sustainability based on use case assessments. Sustain 12, 1–25
(2020). https://doi.org/10.3390/su12176989
33. Hao, Z., Liu, C., Goh, M.: Determining the effects of lean production and servitization of
manufacturing on sustainable performance. Sustain. Prod. Consum. 25, 374–389 (2021).
https://doi.org/10.1016/j.spc.2020.11.018
34. Abdelkafi, N., Pero, M., Masi, A., Capurso, I.: Revisiting the servitization-sustainability link:
a case study in the professional printing supply chain. Clean. Logist. Supply Chain 4, 1–13
(2022). https://doi.org/10.1016/j.clscn.2022.100061
35. Powell, D., Strandhagen, J.O., Tommelein, I., Ballard, G., Rossi, M.: A new set of principles
for pursuing the lean ideal in engineer-to-order manufacturers. Procedia CIRP 17, 571–576
(2014). https://doi.org/10.1016/j.procir.2014.01.137
36. Dallasega, P., Rojas, R.A., Bruno, G., Rauch, E.: An agile scheduling and control approach
in ETO construction supply chains. Comput. Ind. 112, 1–12 (2019). https://doi.org/10.1016/
j.compind.2019.08.003
37. Masi, A., Pero, M., Abdelkafi, N.: Supply chain antecedents of servitization: a study in ETO
machinery companies. Int. J. Prod. Econ. 258, 1–13 (2023). https://doi.org/10.1016/j.ijpe.
2023.108808
38. UCIMU: UCIMU presents the first sustainability report of the machine tool sector
(2023). https://www.ucimu.it/en/press/press-releases/read/2023/02/ucimu-presents-the-first-
sustainability-report-of-the-machine-tool-sector/. Accessed 19 Apr 2023
39. Chipulu, M., Ikeatuegwu, C.: Investigating cultural and structural influences on optimizing
engineer-to-order project delivery and sustainability. IEEE Trans. Eng. Manage., 1–39 (2022).
https://doi.org/10.1109/TEM.2022.3213856
 Investigating On-Site Production
in Construction Using Decoupling Thinking

Petter Haglund1(B) , Joakim Wikner2 , and Martin Rudberg1

1 Department of Science and Technology, Linköping University, Norrköping, Sweden
{petter.haglund,martin.rudberg}@liu.se
2 Department of Production Economics, Linköping University, Linköping, Sweden

[email protected]

Abstract. On-site activities are typically performed when the end-product is

immovable once it has been assembled and/or due to weight and space constraints
in the production and transportation system. It is primarily a customer order driven
type of production with a high level of customization. However, some customer
order driven and/or customization activities can, or should, be performed off-
site, but these aspects are typically treated in isolation of each other. There is
therefore a need to make a combined assessment of on-site activities, customer
order driven production, and customization activities. The purpose of this paper
is thus to develop a typology that provides a holistic view of on-site activities
in relation to customer order driven and customization activities. A flow-based
three-dimensional typology is developed to assess combinations of three dimen-
sions: flow driver, flow differentiation, and flow location. The property of each
dimension is determined using decoupling points. The typology represents how
on-site activities impact the feasibility of customer order driven and customiza-
tion activities. “On-site” is defined as pursuing activities at the site where the
final delivery is made. The findings show that forecast driven activities are risky
to perform on-site and these should instead be performed before delivery to the
site (off-site). The typology is applied to three case companies to illustrate its
applicability to building construction. Future studies should extend the typology
application to other types of construction and ETO industries to enable bench-
marking of site-based production to other contexts with similar challenges related
to on-site/off-site production.

Keywords: Decoupling Point · Engineer-to-Order · Fixed-Position Layout · Site

Production · Building Construction

1 Introduction
Construction always carries some element of on-site production because the final prod-
uct is too heavy and too large to move when it has been assembled [1]. It is therefore
necessary to ensure that on-site activities can be performed as efficiently as possible.
While on-site production typically is associated with construction, its underlying prin-
ciples can be extended to other ETO industries as well. By learning from practices used
© IFIP International Federation for Information Processing 2023
Published by Springer Nature Switzerland AG 2023
E. Alfnes et al. (Eds.): APMS 2023, IFIP AICT 691, pp. 126–139, 2023.
https://doi.org/10.1007/978-3-031-43670-3_9
 Investigating On-Site Production in Construction 127

in on-site production in construction, industries such as aerospace, shipbuilding, and

energy can improve their production processes. Just as in construction, these sectors can
perform the final assembly and installation on-site. Moreover, by using of prefabrication,
these industries can identify opportunities to enhance efficiency by performing certain
activities upstream, and off-site, in the supply chain [2]. Here the construction industry
serves as an example of site-based production, and how it can be used for benchmarking
site-based production to other ETO industries.
ETO is a type of system where not only production but also engineering activities are
performed to customer orders [3]. This type of system means that much of the customer
order driven flow is also customized for the customer order [4]. ETO therefore differs
from make-to-order, which refers to customer order driven production of an already
existing product [5]. Besides performing engineering activities based on customer orders,
several ETO industries perform the final assembly of the product where it is also finally
delivered to the customer, i.e., at a delivery site. This is referred to as “on-site production”
[1]. On-site production is usually, but not necessarily, organized as a fixed position layout,
where production resources (materials, machines, and workers) are moved to the site
where the final installation and assembly of the end-product are performed [6]. However,
the definition of the fixed position layout is a simplification that disregards value-adding
activities that take place prior to the final assembly (i.e., prefabrication). For instance,
in construction, aerospace, shipbuilding, and the energy industry, some activities can be
performed more efficiently in a controlled factory environment, i.e., the supply site(s)
located upstream1 in the value chain of the delivery site. This resembles a more traditional
production system in which the product is moved between fixed production resources,
in contrast to the delivery site where resources are moved to the site. There is thereby
a need to differentiate between activities performed at the supply site(s) and a delivery
site.
Using the concepts of delivery site and supply site(s), the flow is divided into two
parts based on the flow location, i.e., where the flow of activities is located: value-adding
activities that take place at the delivery site (on-site) and activities performed at a supply
site(s), typically in a factory (off-site). In contrast to on-site activities, off-site activities
are typically performed in another type of production system, e.g., job shop, batch flow,
or production line, but they can also be performed using fixed-position layout (e.g.,
assembly islands for heavy machinery that are performed inside a factory). Traditional
construction works is a typical form of on-site production (i.e., where at least the final
assembly is performed at a customer order unique site) combined with fixed position
layout [1].
The combination of on-site and off-site activities can be seen as a form of hybrid
production system where the final assembly at the delivery site is preceded by another
type of production system at the supply site(s). In these hybrid production systems, it

1 “Upstream” and the related term “downstream” refer to the separation of material flow into
the flow towards and away from an organization, production network node, or similar. The
assumption is that the flow is one-directional, but this is not necessarily the case in every
situation (the customer can sometimes become the supplier, e.g., in circular supply chains
or for reverse material flows). However, for the sake of simplicity, this paper only takes the
one-directional material flow into consideration.
128 P. Haglund et al.

is necessary to manage production at multiple sites holistically. As such, it resembles a

form of multi-site production structure where the demand for prefabricated components
in a supply site(s) (i.e., off-site) location is triggered by production at the delivery site
(i.e., on-site). The two types of sites can have different resource configurations depend-
ing on what type of activities that are performed at the respective sites (e.g., different
layouts, equipment, or machinery), but at a minimum the final installation and assem-
bly is performed on-site. Sometimes the product characteristics requires more activities
to be performed on-site (e.g., due to weight or size that makes the product difficult to
move once assembled) and thus a larger part of the customer order driven flow must be
performed at the delivery site.
The differences between on-site and off-site activities lead to challenges in terms of
combining multiple types of production planning and control methods. The selection of
production planning and control methods are typically based on the level of repetitiveness
[7]. In many cases, this means that only one type of planning and control approach is not
appropriate for hybrid production systems since off-site activities are characterized by a
high level of repetitiveness and on-site are characterized by a low level of repetitiveness.
Instead, it can be necessary to combine a batchwise production planning and control
mode (e.g., MRPII) for off-site activities and a one-off production planning and control
mode (e.g., critical path method) for on-site activities.
Alongside the decision of whether to perform activities on-site or off-site, it is neces-
sary to decide when the customer order should enter the flow (i.e., related to flow driver)
and what part of the customer order driven flow that is customized (i.e., related to flow
differentiator). Here it is crucial to assess whether all on-site activities are possible to
perform to a customer order and whether all on-site activities can be customized. For
instance, it can be a risky strategy to perform forecast-driven activities (either standard-
ized or customized) at the delivery site. Companies that operate under the ETO type
production with some element of on-site production therefore face three main decisions:
1) when the customer order (should) enter the flow (flow driver), 2) what part of the
customer order driven flow that is also customized (flow differentiator), and 3) where in
the flow that activities are performed at the delivery site (flow location).
The purpose of this paper is to develop a typology that provides a holistic view of
on-site activities in relation to customer order driven and customization activities. The
typology can provide guidance for how to manage a combination of off-site and on-site
activities more efficiently by applying appropriate planning and control methods given
the planning conditions on-site and off-site, respectively. The typology can be used for
cross-industry benchmarking to facilitate learning across industries that face similar
challenges related to site-based production. However, in this paper, the application of
the typology is limited to the construction of buildings with a moderate to high level of
off-site production.
The paper is structured as follows. Next the method behind the study is presented,
including how the typology was developed and how the cases were selected. In the fol-
lowing section, the typology dimensions (referred to as taxonomies) are first outlined
before they are combined into two-dimensional typologies. The two-dimensional typolo-
gies are then used to create the three-dimensional typology. The typology is then applied
to three cases to illustrate is applicability on three building contractors. This is followed
 Investigating On-Site Production in Construction 129

by discussion of the paper results and findings, and conclusions with suggestions for
further research.

2 Method
The purpose is fulfilled through a combination of conceptual and empirical research.
The conceptual part of the research includes a review of literature to develop three one-
dimensional taxonomies comprising flow driver, flow differentiation, and flow location
into a three-dimensional typology. Typologies, which comprise of two or more dimen-
sions, are used when taxonomies are insufficient to explain a phenomenon [8]. For
this paper, two- and three-dimensional typologies are developed by combining three
one-dimensional taxonomies.
The empirical part of the research contains three case examples (see Table 8). The
cases have different characteristics within the three dimensions and were selected to
explain the differences between building contractors’ using the typology, i.e., on the
basis of theoretical replication [9]. To test the typology, the case examples included in
this study have at least some degree of off-site production to make a clear illustration of
flow location.
The selected cases are companies within the building construction industry where
there are clear challenges related to on-site and off-site production. Furthermore, the
study was limited to building construction to exemplify the typology rather than to
draw conclusions based on the cases. The typology can be applied to other types of
construction with little or no prefabrication (e.g., traditional craft production) and to
other ETO industries with similar challenges related to on-site and off-site production.
However, in this paper the focus was on building construction since this enabled an
easier comparison of the cases and how they could illuminate the typology. Other types
of construction and ETO industries are left out for future studies.

3 Typology Dimensions
This section provides a review of three flow dimensions for flow design.

3.1 Flow Driver

The flow driver indicates whether activities are performed based on customer order or
forecast. The Customer Order Decoupling Point (CODP) is an established term in oper-
ations management and denotes where in the flow that value-adding activities are driven
by customer order by decoupling forecast driven and customer order driven activities
[10]. The positioning of the CODP is determined by the relationship between the supply
lead time (S) (how long time it takes to supply the product, which can include both pro-
duction and engineering activities) and the delivery lead time (D) [11]. When S is longer
than D, some supply activities will need to be performed to forecast. If S is much longer
than D, this will result in a make-to-stock or a assemble-to-order (depending on how
much longer S is relative to D) situation since almost all activities need to be performed
to forecast. When the D is equal to or longer than the S, all activities can be performed
based on a customer order. Table 1 shows examples of the D:S-relation.
130 P. Haglund et al.

Table 1. D:S-relation (based on [11]).

D:S-relation Example
D << S Make-to-stock (MTS)
D<S Assemble-to-Order (ATO)
D≥S Make-to-Order (MTO) or Engineer-to-Order (ETO) depending on whether
engineering activities are included or not

3.2 Flow Differentiator

The term flow differentiator denotes where in the production flow activities go from
being market or customer generic (i.e., standardized) to customer order unique (i.e.,
customized). To separate these two states, the term Customer Adaptation Decoupling
Point (CADP) is used. The positioning of the CADP is determined by the relationship
between the adaption lead time (A) and the required delivery lead time (D) [11]. In
other words, the relationship between A and D denotes how much of D that is customer
order unique. A longer A than D should be avoided since it means that some forecast
driven activities are also customized. On the other hand, most order driven activities can
be standardized when D is much longer than A. Some order driven activities may also
be standardized when D is slightly longer than A. When the two are equally long, the
customer order driven activities are also customized. Table 2 shows some examples of
the A:D-relation.

Table 2. A:D-relation (based on [11]).

A:D-relation Example
A << D Almost all activities are standardized (customer generic)
A=D Customer order driven activities are also customized
A>D Some forecast driven activities are customized (involves a high risk and should
therefore be avoided)

3.3 Flow Location

Flow location is concerned with the location of activities that are either performed at
the site where the final delivery is made to or at a facility that is positioned upstream of
this delivery site, i.e., the supply site(s). These two states are separated by the Delivery
Site Decoupling Point (DSDP), which denotes the relation between the lead time for
delivery site activities (L) and the delivery lead time (D). In site-based production (e.g.,
construction), L is longer than zero, while L is approximately equal to zero for pure
distribution services with very little value-adding on-site.
In ETO fixed position production, it is common that the product remains in one
position and resources are moved to the product. This location is typically specified
 Investigating On-Site Production in Construction 131

by the delivery site. The DSDP denotes where in the flow activities are performed at
the delivery site. The remaining activities that do not take place at the delivery site are
performed at supply site(s). The DSDP indicates where the flow goes from the supply
site(s) to the delivery site. For instance, for a building contractor the DSDP determines
when in the flow that the value-adding activities are performed at the construction site.
It thereby complements related concepts in construction literature that are based on the
degree of value adding that is performed off-site [12]. The position of the DSDP is
instead based on lead times to enable an assessment of whether order driven activities
are performed at the delivery site (on-site) or at the supply site(s) (off-site).
A much longer D than L means that the final assembly on-site is very short relative
to the delivery lead time. This is typically uncommon in building construction but can be
the case for smaller buildings with a low degree of customization, such as prefabricated
single-family dwellings that are assembled in one or a few days. When the D is longer
than L, a larger share of the delivery lead time is spent on-site, although there is still a
considerable amount of the delivery lead time that is allocated to an off-site location.
When D is equal to L, all order driven activities are performed on-site with no off-
site production. In some cases, it can be necessary to perform some activities at the
delivery site based on forecast (i.e., L is longer than D). In non-site-based production, L
is approximately equal to zero. This means that the product is finished once it arrives at
the delivery location and there are no finishing assembly works performed at the delivery
site. Here there are no elements of site-based production, but it might include very short
on-site activities (e.g., replenishing goods on the customer’s shelves). Table 3 shows
some examples of the L:D-relation.

Table 3. L:D-relation.

L:D-relation Example
L << D Most of the customer order driven flow is performed at the supply site with a
very short assembly time at the delivery site
L<D The customer order driven flow takes place both at the delivery site and supply
site(s)
L=D The entire customer order driven flow is performed at the delivery site
L>D Some forecast driven activities need to be performed at the delivery site
L≈0 The lead time at the delivery site is approximately equal to zero, which means
that almost no value-adding activities (e.g., assembly) are performed at the
delivery site

4 Typology Development
This section presents the typology development comprising the two-dimensional typolo-
gies and the three-dimensional typology. The overall approach taken in this section is
based on decoupling thinking used in previous research [11, 13]. First, the taxonomies
132 P. Haglund et al.

are combined into two-dimensional typologies before the final three-dimensional typol-
ogy is developed. Each taxonomy has two states, where the flow driver is either forecast
driven (FD) or customer order driven (CD), the flow differentiator is either Standardized
(St) or Customized (Cu), and the flow location is either Off-Site (OffS) or On-Site (OnS).

4.1 Two-Dimensional Typologies

Typology 1: Flow differentiator and flow driver

This typology has previously been described by Wikner, Bäckstrand [4]. Table 4 rep-
resents the three feasible combinations the CADP and the CODP (FD-St, CD-St, and
CD-Cu) and one unfeasible combination (FD-Cu). The combination FD-Cu is not fea-
sible because its combines forecast driven production with customization [11]. This is a
high-risk strategy that should be avoided.

Table 4. Combined flow differentiator and flow driver.

Forecast driven (FD) Customer order driven (CD)

Customized (Cu) Infeasible (high risk) Feasible (low risk)
Standardized (St) Feasible (low risk) Feasible (low risk)

Typology 2: Flow driver and flow location

Table 5 contains three feasible combinations with low risk (OffS-FD, OnS-CD, and
OffS-CD) and one feasible, but potentially risky, combination (OnS-FD). The situation
OnS-FD is feasible but comes with some risk since forecast driven activities would need
to be performed at the delivery site. However, this can be a feasible option, e.g., for a
developer that has acquired land in an attractive area and decides to build houses since
they are confident that they will be able to sell the houses. In other cases, it is risky
because once the final assembly is completed at the site, the product is difficult to move
(and if the location does not suit the customer’s needs, the supplier will not be able to
sell the product).

Table 5. Combined flow driver and flow location.

Forecast driven (FD) Customer order driven (CD)

Off-site (OffS) Feasible (low risk) Feasible (low risk)
On-site (OnS) Feasible (medium risk) Feasible (low risk)

Typology 3: Flow differentiator and flow location

Table 6 contains three feasible combinations with low risk (OnS-St, OnS-Cu, and OffS-
St) and one feasible combination with a medium risk level (OffS-Cu). Although all four
 Investigating On-Site Production in Construction 133

combinations are feasible, it can be risky to perform some customization activities at

the supply site(s). This is most often dependent on product’s physical properties that are
related customization. For instance, consider an end-product that contains standardized
sub-assemblies that can easily be transported when they are not assembled, but when
assembled the end-product is too large to transport to the customer. For this product, it is
necessary to pursue customization activities at the delivery site due to the physical prop-
erties of the assembled end-product. In the same example, but where the sub-assemblies
are also customized, it is possible to perform customization activities at the supply site(s).
It can also be related to import tolls, where the toll for components and sub-assemblies
is lower than for end-products, which makes it more viable to ship components and
sub-assemblies to the delivery site where they are assembled.
Therefore, the combination OffS-Cu is thus a medium-risk strategy that depends on
the physical properties of the product when it is customized and whether it is beneficial
to pre-assemble the end-product before it is delivered to the delivery site considering
constraints in the production system, transportation, materials handling, etc. Further-
more, it can be disadvantageous to perform customization activities at the supply site(s)
since this requires the off-site production system to cope with high process variability.
A high degree of customization involves many changeovers and small production runs
in combination with high investment costs for off-site production facilities [2].

Table 6. Combined flow differentiator and flow location.

On-site (OnS) Off-site (OffS)

Customized (Cu) Feasible (low risk) Feasible (medium risk)
Standardized (St) Feasible (low risk) Feasible (low risk)

4.2 Three-Dimensional Typology

The three two-dimensional typologies each represent how one decision within each one
of the three dimensions (flow differentiator, flow driver, and flow location) affects the
other. However, building contractors need to make a combined assessment of decisions
within the three dimensions when designing their production system. The three dimen-
sions have been combined in Table 7. By combining the three dimensions, we get a total
of eight states/combinations where four are feasible (low risk), two are feasible with a
medium risk level, and two are infeasible (high risk).
This typology can be used to outline recommendations concerning appropriate pro-
duction methods, investments in off-site facilities, guidelines for design and engineering
activities (e.g., what level of customization to offer clients), and to implement appropriate
planning and control methods. The eight combinations in the typology represent differ-
ent feasible and unfeasible states that result from how the CADP, CODP, and DSDP
are positioned in relation to each other. The next section contains applications of the
three-dimensional typology to illustrate how the typology can be used to assess existing
configurations.
134 P. Haglund et al.

Table 7. Combined flow driver, flow differentiator, and flow location.

# Flow driver Flow Flow Feasibility (risk level)

differentiator location
1 Forecast driven Standard On-site Feasible (medium risk)
2 Forecast driven Standard Off-site Feasible (low risk)
3 Forecast driven Customized On-site Infeasible (high risk)
4 Forecast driven Customized Off-site Infeasible (high risk)
5 Customer order driven Standard On-site Feasible (low risk)
6 Customer order driven Standard Off-site Feasible (low risk)
7 Customer order driven Customized On-site Feasible (low risk)
8 Customer order driven Customized Off-site Feasible (medium risk)

5 Typology Application
In this section, the three-dimensional typology is applied to three case companies within
building construction. Although the taxonomies and the two-dimensional typologies can
be used to analyze one or two aspects of the case companies’ production system, for
the purpose of this paper, the focus is on the three-dimensional typology. Each case is
analyzed by positioning the CADP, CODP, and DSDP in relation to each other in their
respective production systems. Table 8 shows the cases’ lead time relations for flow
driver, flow differentiator, and flow location.

Table 8. Case applications.

Case D:S-relation A:D-relation L:D-relation

1 ETO: Fully customized: A = D Combination of on-site and off-site:
D≥S L<D
2 ETO: Combination of standardized and Combination of on-site and off-site:
D≥S customized: A < D L<D
3 ETO: Fully customized: A = D Combination of on-site and off-site:
D≥S L<D

5.1 Case 1
The production process contains a relatively high degree of off-site production, but
the building system is still flexible. They prefabricate panelized elements with pre-
assemblies (windows, doors, radiators, and electrical fittings). Floor slabs are also com-
pleted with parquet flooring and the roof is delivered to the site complete and ready
for assembly. The final assemblies of the panelized elements, floor slabs, and the roof
 Investigating On-Site Production in Construction 135

are then performed on the construction site along with assembly of non-load bearing
components and assemblies (e.g., interior walls).
The S is longer than the L and the DSDP is positioned between the production of
the prefabricated panelized elements and final assembly works. The S is equal to the D
(where S includes engineering activities), which means that the building system operates
under an ETO basis. All customization activities are performed to customer order. For
instance, each individual wall, floor slab, and roof is tailored based on the architectural
design. Hence, the entire D is customized (i.e., D and A are equally long).
The CADP and the CODP coincide with each other, since all order driven activ-
ities are customized, and are positioned upstream of the DSDP. A closer look at the
CODP shows that it is positioned between the production/sourcing of raw materials and
the prefabrication of panelized elements. Panelized elements are produced to customer
order and are customized for each project. The design and engineering tasks and value-
adding activities in the off-site factory are thus customer order driven to a high degree to
accommodate a flexible market segment in combination with industrialized construction
methods. Thus, the building system can utilize the benefits of using an off-site produc-
tion facility to make customized products, although with a medium risk level since it
can be difficult to reach the breakeven production volume to offset the off-site facilities
investment costs.

5.2 Case 2

The building system uses seven standardized volumetric modules that are combined into
customized product variants. The volumetric modules are complete functional spaces that
enclose bathroom/entrance, kitchen, or bedroom. They are also installed with electrical
fittings, plumbing, façade material, windows, doors, etc. When the volumetric modules
have been assembled in the factory, they are transported to the construction site for final
assembly. Here the contractor supplies elevators, exterior corridors, and balconies that
are delivered from the supplier directly to the construction site.
The D is longer than the L and the DSDP is positioned between the volumetric module
assembly and the final assembly. The volumetric modules are however still produced
to customer order (S is equal to D). Because some design and engineering activities
are standardized (the design of the volumetric modules), the D is moderately longer
than the A. This means that some order driven activities are customized, and some are
standardized.
Because the production is customer order driven (MTO), the CODP is positioned
upstream of both the DSDP and CADP. The company only speculates on an inventory of
raw materials and components. Therefore, the CODP is positioned between the produc-
tion/sourcing of raw materials and components and the production of panelized elements
and floor slabs. Downstream of the CODP, the CADP and the DSDP coincide and are
positioned in between the volumetric module assembly in the off-site factory and the
final assembly on site.
136 P. Haglund et al.

5.3 Case 3
In case 3, production of volumetric modules is performed to customer order in an off-site
factory. The volumetric modules are then assembled on-site along with final assembly
of interior walls, windows, doors, façade, and the roof. The D longer than the L and the
DSDP is positioned between the volumetric module assembly and the final assembly.
The final assembly does however comprise more components and assemblies than in
case 2, where only the assembly of volumetric modules remained on-site. As with the
two previous cases, case 3 is an ETO type (S is equal to D).
Regarding product customization, the building system comprises volumetric mod-
ules that are constrained in dimensions (maximum 9,6m x 3,7m) to accommodate the
dimensions of the production line in the off-site factory and the trucks or boats for trans-
portation. Other than the dimensions of the volumetric modules, architects and engineers
have only minor constraints related to size of openings, shape of the volumetric modules,
etc. The volumetric modules are therefore customized for each project. This means that
the D is equal to the A since each volumetric module must be designed from scratch in
each project.
The CODP and CADP coincide and are positioned between the production/sourcing
of raw materials and components and the volumetric module assembly, whereas the
DSDP is positioned further downstream in between volumetric modules assembly and
final assembly on-site. Both design and engineering tasks and production are per-
formed to customer order. The volumetric modules can therefore be adapted to each
project’s unique circumstances, which comes at the expense of more extensive design
and engineering work for each customer order. Furthermore, the off-site factory needs
to accommodate a larger variation in volumetric module designs compared to case 2.

6 Discussion
The three cases illustrate the effects of how the positioning of the CADP, CODP, and the
DSDP relative to each other can be used for flow design based on decoupling points. The
position of the CADP determines the level of product customization. This is affected by
the time required to make customer order specific adaptions in design and engineering
(A) in relation to the required delivery lead time (D) by the customer. As such, CADP
should not be positioned upstream of the CODP since this would mean that the product
is customized based on forecast. Typically, this is also goes for the DSDP. On-site and
forecast driven production can potentially be risky, although it is possible to perform
activities at the delivery site to forecast. As seen in the cases, none of the contractors
pursued forecast driven activities on-site. However, the DSDP can be positioned both
upstream and downstream of the CADP. By positioning the DSP downstream of the
CADP, the customized flow takes place off-site. Although this is possible, it should be
carefully assessed since it can be difficult to achieve depending on physical properties
of the product and the capabilities of the production and transportation system.
As seen in the cases, by analyzing the relative positions of the CODP, CADP, and
DSDP in the flow, it is possible to determine feasible levels of prefabrication. However,
it is important to note the difference between the lead time-based concepts used in this
paper in contrast to the value-adding based concepts typically considered in literature.
 Investigating On-Site Production in Construction 137

Certain combinations (see Table 7) of flow driver, flow differentiation, and flow location
involves risks. The typology can therefore be used to identify where to allocate resources
to mitigate risks and where to place time and/or inventory buffers. For example, the DSDP
can be used as a point in the flow to decouple the flow using a time or inventory buffer
since it can be difficult to synch off-site production with on-site production. It can also
be used as a lot sizing buffer, but this depends on where the CODP, CADP, and DSDP
are positioned relative to each other in the flow. If the CADP is positioned upstream
of the DSDP, it is not recommended to have a lot-sizing buffer positioned at the DSDP
because the CADP is the last point in the flow that this type of buffer should occur
[11]. Instead, the buffer positioned at the DSDP will be used solely for the purpose of
balancing off-site and on-site activities (i.e., an off-site buffer) rather than for balancing
inventory holding costs and setup costs at the supply site(s).
Furthermore, the typology presented in Table 7 can be used as a reference for imple-
menting planning and control methods that are suitable given the risk level, product
customization, and degree of off-site production. For instance, in combination 2 (fore-
cast driven, standard, off-site) where production of a standard product is performed to
forecast off-site, it is likely better to use intermittent planning and control approaches
suitable for batchwise transformation to reduce the total amount of set-up time. How-
ever, in combination 7 (customer order driven, customized, on-site), it is likely better
to pursue planning and control methods suitable for one-off production (e.g., using the
critical path method).
Typically, site-based production has a low level of repetitiveness, which is in favor of
one-off production mode planning and control approaches, while off-site production has
a higher level of repetitiveness, which is in favor of intermittent planning and control
approaches [7]. Case 2 and 3 had positioned the DSDP after the volumetric module
assembly, whereas in case 1, the company had positioned the DSDP after the panelized
elements and floor slabs were finished. Case 2 and 3 can therefore perform a larger share
of production (in terms of lead times) in a controlled factory environment with a higher
level of repetitiveness, which suits batchwise planning and control methods. However,
final assembly of the volumetric modules along with final construction works needed
to be performed on-site in a more traditional manner with onetime mode planning and
control approaches. Case 1 had a larger share of the production lead time on-site, which
means that the use of a one-off production mode planning and control method will be
more prominent than in case 2 and 3, which uses line/batch flow in the off-site production
facility.
The main challenge, however, will be to combine different modes of planning and
control when the production system consists of both off-site production and on-site pro-
duction. Incorrect sequencing and poor coordination from off-site production is common
in building construction that combine off-site and on-site production [14]. To ensure both
efficient and effective off-site production and on-site production, one-off production
mode methods typically used in on-site production planning will need to be coordinated
with the batchwise production components and assemblies in the factory. The mass-
production oriented logic used in off-site production facilities and in transportation do
not cater the requirements for organizing shipments to the construction site according
to assembly sequences [15]. The complexity in coordinating these activities is further
138 P. Haglund et al.

increased if the contractor is not in control of the off-site production. This often occurs
in construction, e.g., when the contractor buys windows that are supplied directly to the
construction site from the supplier.

7 Conclusions and Further Research

The purpose of this paper is to develop a typology that provides a holistic view of on-site
activities in relation to customer order driven and customization activities. Decoupling
thinking was used to distinguish between different dimensions related to the flow driver,
flow differentiation, and flow location. The three dimensions represent three impor-
tant strategic decisions for companies that pursue site-based production to some extent.
The two-dimensional and three-dimensional typologies illustrates the interdependence
between the three dimensions which need to be considered holistically to ensure effi-
cient and effective operations. The cases illustrate the importance of making a holistic
assessment of the relative positions of the CODP, CADP, and DSDP. Furthermore, the
typology can be used to position time and inventory buffers and to select appropriate
planning and control methods.
This paper extends prior research by focusing on strategic decision areas that relate
to challenges in flow design in site-based production systems. This paper has focused
on building contractors with moderate to high degrees of off-site production, but the
typology can be applied to construction using traditional craft production as well as
other ETO contexts. The typology is developed with other types of site-based production
systems in consideration and further studies should apply the three-dimensional typology
to other cases to test its generalizability. Finally, this paper only considers the customer’s
requirements from the perspective of the contractor. Future studies should investigate the
supply perspective, and perhaps also the demand perspective, and include the flow that
is external to the contractor, i.e., extend the system boundaries of the flow delimitation
[11].

References
1. Ballard, G., Howell, G.: What kind of production is construction. In: 6th Annual Conference
International Group for Lean Construction, pp. 13–15 (1998)
2. Jonsson, H., Rudberg, M.: Classification of production systems for industrialized building: a
production strategy perspective. Constr. Manag. Econ. 32(1–2), 53–69 (2014)
3. Wortmann, J.C., Muntslag, D.R., Timmermans, P.J.: Customer-driven Manufacturing.
Springer (1997). https://doi.org/10.1007/978-94-009-0075-2
4. Wikner, J., Bäckstrand, J.: Triadic perspective on customization and supplier interaction in
customer-driven manufacturing. Prod. Manuf. Res. 6(1), 3–25 (2018)
5. Wikner, J., Rudberg, M.: Integrating production and engineering perspectives on the customer
order decoupling point. Int. J. Oper. Prod. Manag. 25(7), 623–641 (2005)
6. Hill, A., Hill, T.: Manufacturing operations strategy. Palgrave Macmillan (2009)
7. Wikner, J.: An ontology for flow thinking based on decoupling points–unravelling a control
logic for lean thinking. Prod. Manuf. Res. 6(1), 433–469 (2018)
8. Meredith, J.: Theory building through conceptual methods. Int. J. Oper. Prod. Manag. 13(5),
3–11 (1993)
 Investigating On-Site Production in Construction 139

9. Eisenhardt, K.M.: Building theories from case study research. Acad. Manag. Rev. 14(4),
532–550 (1989)
10. Hoekstra, S., Romme, J.: Integral Logistic Structures: Developing Customer-Oriented Goods
Flow. McGraw-Hill, London (1992)
11. Wikner, J.: On decoupling points and decoupling zones. Prod. Manuf. Res. 2(1), 167–215
(2014)
12. Jonsson, H., Rudberg, M.: Production system classification matrix: matching product
standardization and production-system design. J. Constr. Eng. Manag. 141(6), 05015004
(2015)
13. Wikner, J., Noroozi, S.: A modularised typology for flow design based on decoupling points–
a holistic view on process industries and discrete manufacturing industries. Prod. Plann.
Control. 27(16), 1344–1355 (2016)
14. Pan, W., Parker, D., Pan, M.: Problematic interfaces and prevention strategies in modular
construction. J. Manag. Eng. 39(2), 05023001 (2023)
15. Bortolini, R., Formoso, C.T., Viana, D.D.: Site logistics planning and control for engineer-to-
order prefabricated building systems using BIM 4D modeling. Autom. Constr. 98, 248–264
(2019). https://doi.org/10.1016/j.autcon.2018.11.031
 Clarifying the Interface Between Construction
Supply Chain and Site - A Key to Improved
Delivery Efficiency

Farah Naz1,2(B) and Anna Fredriksson1

1 Linkoping University, Campus, 60174 Norrköping, Sweden
{I.farah.naz,II.anna.fredriksson}@liu.se
2 The Swedish National Road and Transport Research Institute (VTI), 58330 Linköping,

Sweden

Abstract. The purpose of the paper is to capture non-value adding activities of

the order-to-delivery process at the interface between construction site and the
supply chain to identify efficiency improvement potentials. The paper is based
on a case study of a Swedish construction site. The unit of observation has been
the activities taking place at the interface between construction supply chain and
construction site. i.e. deliveries and on-site material handling. Data is collected
through observations in the unloading zone of the construction site, unstructured
interviews, and a workshop with site management. The findings of the paper show
that there exists fuzziness in the interface of construction supply chain and con-
struction site causing the observed non-value adding activities such as waiting,
excessive material handling, unnecessary vehicle movements etc. The root causes
of the fuzziness are poor planning, lack of communication between site manage-
ment and transporter/material supplier, and lack of standard operating procedures
for unloading. The paper provides knowledge of the importance of managing
interface between construction supply chain and site to remove non-value adding
activities. Additionally, the paper provides a clear view of the different actors’
roles in the non-value adding activities and give suggestions of how to improve
efficiency of construction deliveries. The paper is based on a single case study and
further studies are needed to provide deeper insights into the phenomenon.

Keywords: Construction Supply Chain · Logistics Planning and Control ·

Non-Value Adding Activities

1 Introduction

Construction accounts for 13% of gross domestic product (GDP) globally and employs
7% of the world´s working population [6]. Urbanization and modern living have further
increased the importance of this sector [35]. Construction industry is recognized as an
engineer-to-order (ETO) industry, given that the construction products (i.e. buildings)
are physically big and immobile, are produced at the site of use and are customized

© IFIP International Federation for Information Processing 2023

Published by Springer Nature Switzerland AG 2023
E. Alfnes et al. (Eds.): APMS 2023, IFIP AICT 691, pp. 140–153, 2023.
https://doi.org/10.1007/978-3-031-43670-3_10
 Clarifying the Interface Between Construction Supply Chain and Site 141

to meet the specific needs and requirements of clients [8]. The construction process
is typically carried out within temporary organizations, requiring the establishment of
temporary supply chains [32] where lo