Integration and Optimisation of an RFID-Enabled

Inventory Management System of a Future

Generation Warehousing System

by
SALEH AL-YAHYA

Thesis for the Degree of Doctor of Philosophy

September 2017
Abstract

It is widely accepted that future generation warehouses require a real-time visibility

and accuracy of inventory data in order to maintain efficient and effective warehousing
operations, optimal SKU levels, and up-to-date inventory management and control of
incoming and outgoing goods that often occur today at increasingly centralised
distribution centres. This phenomenon is partially due to a sharp rise of online
shopping activities in many countries where customers now prefer to purchase goods
online and demand a fast delivery of ordered products to be dispatched directly at their
door steps. Thus, there is a strong desire from supply chain and logistics sectors to
seek even more efficient and cost-effective methods for sorting, storing, picking and
dispatching goods at increasingly centralised distribution centres in which automation
and integration of warehousing systems is inevitable. As part of a research programme
for future generation warehouses, this thesis presents an investigation into some
design theories and an integrated optimisation methodology for a future generation
warehousing system in which an RFID-based inventory management system has the
capability of interacting with a proposed RFID-enabled automated storage and
retrieval mechanism without any human intervention. An efficient item selection
algorithm based on pre-defined rules was developed and implemented within an RFID-
based inventory management system, which also allows a manipulation of RFID-
tracked items to seek an optimal solution by assigning a priority to one of selected
items to travel in an order (if applicable) with both minimal travel time and waiting time
to a specified collection point; this maximises efficiency in material-handling and
minimises operational costs. A pilot test was established and carried out based on the
proposed RFID-enabled inventory management system for examining the feasibility
and applicability of its embedded RFID-enabled item-selection optimisation algorithm.
Experimental results demonstrate that the developed methodology can be useful for
determining an optimal solution (or route) for the RFID-enabled pusher to push a
selected RFID-tagged item located randomly from a storage rack to an output
conveyor system in a sequence that allows all the selected items traveling to a
specified collection point with a minimal waiting time for packing. In theory, such a
system can also be expanded by incorporating other pre-defined selection parameters
as requested by users. Moreover, a multi-objective model using the multi-criterion
fuzzy programming approach was developed and used for obtaining a trade-off

i
decision based on conflicting objectives: minimisation of the total cost, maximisation
of capacity utilisation, maximisation of service level and minimisation of travel distance
within the proposed warehousing system. The developed model also supports design
decisions in determining an optimum number of storage racks and collection points
that need be established for the proposed warehouse. To reveal the alternative
Pareto-optimal solutions, a decision-making algorithm namely TOPSIS was also
employed to select the best Pareto-optimal solution obtained using the multi-criterion
fuzzy programming approach. Case-studies were also conducted to demonstrate the
feasibility and applicability of the developed multi-objective model and optimisation
methods. The study concluded that the research work provided a useful basis by
developing a framework as part of contributions in design theories and optimisation
approaches for integration of future generation RFID-based warehousing systems and
a practical means of exploring the further work in this field.

Keywords: Warehouses, Inventory, RFID, Automation, Integration, Optimisation,

Supply chains.

ii
Declaration
This is to certify that I am responsible for the work submitted in this thesis, that the
original work is my own except as specified in acknowledgments or in footnotes, and
that neither the thesis nor the original work contained therein has been submitted to
this or any other institution for a higher degree.

Author's signature ………………………………………………………………

Date ………………………

iii
Publications

Journal

 A. Mohammed, Q. Wang, S. Alyahya, N.Bennett “Design an optimisation of an

RFID-enabled warehousing system under uncertainties: a multi-criterion fuzzy
programming approach”, The International Journal of advanced manufacturing
technology, 91(5), pp.1661-1670, 2016.

 S.Alyahya, Q.Wang, N.Bennett, “Application and integration of an RFID-

enabled warehousing management system - A feasibility study”, Journal of
Industrial Information Integration, Vol. 4, pp. 15-25, 2016.

Conference

 A. Mohammed, S. Alyahya, Q. Wang, “Multi-objective optimisation for an RFID-

enabled automated warehousing system”, IEEE International Conference on
Advanced Intelligent Mechatronics (AIM), pp.1345-1350, Banff, Alberta,
Canada, July 2016.
 S. Alyahya, Q. Wang, “An RFID-embedded mechanism of automated storage
and retrieval racks”, International Conference on Electrical, Electronic and
Computer Engineering Technologies (ICEECET 2016), pp.62-77, Shanghai,
China, April 2016.
 Q. Wang, S. Alyahya, N. Bennett and H. Dhakal, “An RFID-enabled automated
warehousing system”, 2015 International Conference on Robotics, Mechanics
and Mechatronics (ICRMM 2015), Singapore, March 2015; International
Journal of Martials, Mechanics and Manufacturing (IJMMM), Vol. 3, pp. 287-
293, 2015.
 S. Alyahya, Q. Wang, G. Yang, “Mathematical modelling of a U-shaped
assembly cell using flexible workers”, Academic Journal of Art and Sciences’
(IJAS) Conference, Academic Journal of Science, Vol.1, No.2, pp.21-26, Boston,
USA, May 2012.
Research Seminar

 S. Alyahya, “Application and integration of an RFID-enabled warehousing

management system– a feasibility study”, University of Portsmouth, UK, 9th
December 2015.

iv
Abbreviations

IT: Information Technology.

RFID: Radio Frequency Identification.

TX: Transmitter Antenna.
RX: Receiver Antenna.
NFC: Near Field Communication.
FFC: Far Field Communication.
EMF: Electro-Magnetic Fields.
UID: Unique Identification.
PWM: Pulse Width Modulation.
HF: High Frequency.
LF: Low Frequency.
UHF: Ultra High Frequency.
EPC: Electronic Product Code.
RTD: Row Tag Database.
SKUs: Stock Keeping Units.
Auto-ID: Automatic Identification.

WMS: Warehouse Management System.

DWMS: Digitalised Warehouse Management System.

RMS: Resource Management System.

ECA: Event-Condition-Action.

RICMS: RFID-based Inventory Control and Management System.

PLM: Product Life Cycle Management.

ERP: Enterprise Resource Planning.

AIM: Application Implementation Methodology.
WSN: Wireless Sensor Network.
LabVIEW: Laboratory Virtual Instrumentation Engineering Workbench.
LAN: Local Area Network.

v
WLAN: Wireless Local Area Network.
PLC: Programmable Logic Controller.
AS/RS: Automated Storage and Retrieval System.
AS/RR: Automated Storage and Retrieval Rack.
ACO: Ant Colony Optimisation.
VNS: Variable Neighbourhood Search.
BOSC: Build-to-Order Supply Chain.
B2B: Business-to-Business.
RPC: Returnable Produce Container.
AGV: Automated Guided Vehicle.
EOQ: Economic Order Quantity.
ICs: Integrated Circuits.
PET: Polyethylene Therephtalate.
RC: Reader Coordinator
DC: Data Collector.
IDE: Integrated Development Environment.
LLRP: Low Level Reader Protocol.
MVS: Microsoft Visual Studio.
ORM: Object Relational Mapping.
API: Application Programming Interface.
ODBC: Open Database Connectivity.
GUIDE: Graphical User Interface Development Environment.
GUI: Graphical User Interface.
DMS: Database Management System.

vi
Table of Contents
ABSTRACT ................................................................................................................ I
PUBLICATIONS ....................................................................................................... IV
ABBREVIATIONS...................................................................................................... V
LIST OF FIGURES..................................................................................................... X
LIST OF TABLES .................................................................................................... XII
ACKNOWLEDGEMENTS ....................................................................................... XIII
CHAPTER 1 INTRODUCTION .................................................................................. 1
1.1Research background................................................................................................... 1
1.2 Research aim and objectives ....................................................................................... 4
1.3 Thesis organisation ..................................................................................................... 6
CHAPTER 2 INDUSTRIAL SURVEY AND EMPIRICAL ANALYSIS ........................ 7
2.1 Applications of database.............................................................................................. 8
2.2 Warehouse operations................................................................................................. 8
2.3 Inventory management ................................................................................................ 9
2.4 E-commerce services ................................................................................................ 10
2.5 Delivery ..................................................................................................................... 10
Summary ......................................................................................................................... 12
CHAPTER 3 LITERATURE REVIEW ...................................................................... 13
3.1 Inventory management systems ................................................................................ 14
3.1.1 MRP/ERP............................................................................................................ 14
3.1.2 Warehouse management systems (WMS) .......................................................... 15
3.1.3 Lean management .............................................................................................. 16
3.2 Manual warehouse .................................................................................................... 17
3.3 Automated warehouses ............................................................................................. 18
3.4 Internet of things (IoT) ............................................................................................... 19
3.4.1Technology Readiness Levels.............................................................................. 21
3.5 RFID technology and applications ............................................................................. 21
3.5.1 RFID applications in warehousing systems ......................................................... 25
3.6 RFID-based automated warehouse systems ............................................................. 31
Summary ......................................................................................................................... 35
CHAPTER 4 THE RFID-ENABLED INVENTORY MANAGEMENT SYSTEM ........ 36
4.1 RFID tags .................................................................................................................. 36
4.1.1 Passive RFID tags .............................................................................................. 38
4.1.2 Active RFID tags ................................................................................................. 38
4.1.3 Semi-active/Semi-passive RFID tags .................................................................. 39
4.1.4 Selection of RFID tags ........................................................................................ 40
4.2 RFID antennas .......................................................................................................... 41

vii
 4.2.1 Antenna characteristics ....................................................................................... 41
4.3 RFID readers ............................................................................................................. 42
4.3.1 RFID reader performance ................................................................................... 44
4.4 Communication between RFID tags and RFID readers ............................................. 45
4.5 Electromagnetic field ................................................................................................. 46
4.5.1 Near Field communication vs Far Field communication....................................... 46
4.6 RFID controllers......................................................................................................... 47
4.7 RFID software ........................................................................................................... 48
4.7.1 RFID middleware ................................................................................................ 48
4.7.2 Application software and implementation for RFID systems ................................ 49
4.8 RFID cost elements ................................................................................................... 50
Summary ......................................................................................................................... 51
CHAPTER 5 THE PROPOSED RFID-BASED INVENTORY MANAGEMENT
SYSTEM AND MEASUREMENT ............................................................................. 52
5.1 Mechanism of the proposed RFID-based automated warehouse system................... 52
5.2 The RFID-enabled automated storage and retrieval system (AS/RS) ........................ 53
5.3 Measurement of the storage and retrieval mechanism............................................... 55
5.4 Travel time................................................................................................................. 57
5.5 Indicator function ....................................................................................................... 63
Summary ......................................................................................................................... 66
CHAPTER 6 INTEGRATION OF THE RFID-ENABLED WAREHOUSE
MANAGEMENT SYSTEM ....................................................................................... 67
6.1 Key components used for integration of the RFID-enabled warehouse management
system ............................................................................................................................. 67
6.2 RFID near-field communication ................................................................................. 68
6.2.1 Recognition between an RFID tag and an RFID reader....................................... 69
6.3 RFID far-field communication .................................................................................... 71
6.4 Database of the RFID inventory management system ............................................... 73
6.4.1 The data processing algorithm using MATLAB .................................................... 73
6.5 Standardisation of RFID Technology ......................................................................... 75
Summary ......................................................................................................................... 77
CHAPTER 7 A PILOT STUDY AND EXPERIMENTAL RESULTS ......................... 78
7.1 Hardware and software of the RFID inventory management system.......................... 79
7.2 Tests of RFID passive tags ........................................................................................ 83
7.3 RFID tags read rang test ........................................................................................... 85
7.4 RFID application development ................................................................................... 85
7.5 MATLAB simulation results ........................................................................................ 88
Summary ......................................................................................................................... 93

viii
CHAPTER 8 MULTI-OBJECTIVE OPTIMISATION FOR THE RFID-ENABLED
AUTOMATED WAREHOUSING SYSTEM .............................................................. 94
8.1 Problem definition and model description .................................................................. 96
8.2 The optimisation methodology ................................................................................... 99
8.3 Application and evaluation ....................................................................................... 105
8.3.1 Results and discussions .................................................................................... 105
Summary ....................................................................................................................... 110
CHAPTER 9 COST ANALYSIS, BENEFITS AND LIMITATIONS – A REAL CASE
STUDY ................................................................................................................... 111
9.1 Cost estimates ......................................................................................................... 111
9.1.1 Cost analysis for the RFID system vs barcodes ................................................ 112
9.1.2 The interview with project manager ................................................................... 113
9.1.3 The interview method include the questions in such as: .................................... 113
Summary ....................................................................................................................... 116
CHAPTER 10 SUMMARY, DISCUSSIONS, CONCLUSIONS AND FUTURE WORK
............................................................................................................................... 117
10.1 The recommended future work .............................................................................. 122
REFERENCES ....................................................................................................... 123
APPENDIX ............................................................................................................. 137

ix
List of Figures

Figure 1: Share of manual, semi-automated and fully automated warehousing systems. ....... 9
Figure 2: Barcodes, voice picking or other technologies within the warehouse. ..................... 10
Figure 3: Response rate of future generation automated warehouse system. ......................... 11
Figure 4: Typical warehouse functions. .......................................................................................... 13
Figure 5: Seven wastes in lean manufacturing. ............................................................................ 17
Figure 6: Evolution of RFID technology .......................................................................................... 22
Figure 7: RFID applications into different sectors. ........................................................................ 23
Figure 8: Barcodes vs RFID ............................................................................................................. 27
Figure 9: The overall layout of the 3D warehousing system. ...................................................... 32
Figure 10: Operating costs in a typical warehouse. ...................................................................... 34
Figure 11: The key components of the RFID-enabled inventory management system. ......... 36
Figure 12: key components of a typical RFID chip. ...................................................................... 37
Figure 13: A variety of RFID tags .................................................................................................... 37
Figure 14: A typical wave pattern of linear-polarized and circular-polarised antennas (Brown,
2007). ................................................................................................................................................... 42
Figure 15: A typical antenna. ............................................................................................................ 42
Figure 16: A typical stationary reader. ............................................................................................ 43
Figure 17: A typical handheld reader. ............................................................................................. 43
Figure 18: different frequencies between NFC and the FFC. ..................................................... 47
Figure 19: Key components of an RFID-based automated warehouse system. ...................... 53
Figure 20: Structure of the automated storage and retrieval system. ........................................ 54
Figure 21: Random storage and retrieval locations of SKUs. ..................................................... 55
Figure 22: Geometric parameters corresponding to the key components of the storage and
retrieval system. ................................................................................................................................. 56
Figure 23: Geometric parameters in a storage rack. .................................................................... 56
Figure 24: Velocity-time relationship of the travelling S/R pusher for the case I motion where
acceleration, constant speed and deceleration is utilised............................................................ 58
Figure 25: Velocity-time relationship of the S/R pusher travelling- case II................................ 58
Figure 26: Multi-directional trajectory of a pusher which travels to selected items. ................. 62
Figure 27: Mechanism of searching and selecting an ordered item in a storage rack. ........... 64
Figure 28: The routing algorithm for a pusher to travel to each of selected items. .................. 65
Figure 29: Key components used in this experiment. .................................................................. 68
Figure 30: Arduino UNO.................................................................................................................... 69
Figure 31: Reading activities between an RFID tag and an RFID reader. ................................ 70
Figure 32: The RFID Reader (ID-12) .............................................................................................. 71
Figure 33: Key components of the RFID inventory management system for the pilot test. ... 72
Figure 34: The RFID-data processing algorithm. .......................................................................... 74
Figure 35: The prototype of the RFID-inventory management system using an Arduino
interface ............................................................................................................................................... 79
Figure 36: Screen shot of an example of RFID codes generated through the Arduino
platform. ............................................................................................................................................... 79
Figure 37: Screen shot of processed RFID data of an ordered item. ........................................ 81
Figure 38: Screen shot of the constant velocity vs acceleration/deceleration. ......................... 82

x
Figure 39: Testing of the RFID-based inventory management system. .................................... 82
Figure 40: Screen shot of configure the MultiReader software platform ................................... 83
Figure 41: Setting is applied correctly for the MultiReader software platform ......................... 84
Figure 42: Screen shot of RFID codes generated through the MultiReader software platform
.............................................................................................................................................................. 85
Figure 43: MVS window simulator. .................................................................................................. 87
Figure 44: Display of an optimal location of one of a selected item in the GUI. ....................... 88
Figure 45: Determination of a longest travel time among ordered items. ................................. 90
Figure 46: The results of selected items for five different stages with accumulating travel
times. .................................................................................................................................................... 92
Figure 47: Comparison among the obtained objective function values. .................................. 109

xi
List of Tables

Table 1: Qualified responses received from companies. ............................................................... 7

Table 2: Summary of a comparison between barcodes and RFID tags. ................................... 28
Table 3: A number of case studies in RFID applications through a literature review. ............. 29
Table 4: Five common categories of RFID tags. ........................................................................... 40
Table 5: Typical characteristics of RFID tags ................................................................................ 41
Table 6: RFID reader performance parameters. ........................................................................... 44
Table 7: Comparison in performance between NFC and FFC.................................................... 46
Table 8: Specifications of Arduino UNO. ........................................................................................ 69
Table 9: Part of information data within the SQL server. ............................................................. 73
Table 10: Part of MATLAB programming codes for data processing. ...................................... 75
Table 11: RFID components product code. ................................................................................... 75
Table 12: Part of the SDK programming codes for data processing of acquiring RFID tags. 86
Table 13: Determination of an item with a longest travel time in an AS/RR to the collection
point. ..................................................................................................................................................... 89
Table 14: Application data ranges. ................................................................................................ 105
Table 15: Satisfaction degrees for objective functions. .............................................................. 106
Table 16: Score values of Pareto-optimal solution using TOPSIS. .......................................... 107
Table 17: Computational results. ................................................................................................... 108
Table 18: Self-service and security solution cost. ....................................................................... 111
Table 19: Costs for RFID and barcodes at the university of Portsmouth library. ................... 112

xii
Acknowledgements

Several people have supported me throughout my research. The following

acknowledgments go a small way to expressing my thanks. Firstly, I would like to thank
my academic supervisor Dr Qian Wang for his continued guidance and support. He
provided me with the motivation I needed to focus and achieve my goals.

I would like to thank my parents, by far my biggest fans and platinum level supporters,
who through my childhood and study career always encouraged me to follow my heart
and inquisitive mind in any direction this took me. Your support means the most to me.
I love you both.

xiii
Chapter 1 Introduction

1.1 Research background

The traditional concept of automated warehousing systems often refers to applications

of automatic storage and retrieval modules, lifting equipment, robots, AGVs
(automated guided vehicles), conveyor systems and so on for stacking, picking and
transporting incoming and outgoing goods of a distribution centre. These facilities are
used to overcome some disadvantages of manually operated warehouses which often
lead to a high frequency of human errors, a consistent increase of labour costs and
poor efficiency of material-handling operations. In recent years, it has been seen an
exponentially rising number of customers who like ordering products online and expect
a fast delivery of ordered products to be dispatched directly to their door steps.
Because of this new type of online shopping habits, many traditional stores (or
warehouses) are no longer suitable for satisfying such demands of online shoppers
who require the service as described above. A study through a literature review shows
that future generation warehouses may be designed and implemented as more
centralised distribution centres that partly replace conventional stores or warehouses
of manufacturers, suppliers and retailers in supply chain and logistics sectors (Wang
et al., 2010). This requires a novel design of a more efficient and cost-effective
mechanism of automated storage and retrieval systems as a key element of
distribution centres for sorting, storing, picking and dispatching goods. Implementation
and integration of fast-growing IT technologies have demonstrated great improvement
opportunities of a warehouse in terms of a tighter inventory control, a shorter response
time and a greater variety of SKUs (stock keeping units). These capabilities can be
enhanced by using smart-labels such as radio frequency identification (RFID) tags,
automatic identification (Auto-ID) sensors, wireless communication networks and
indoor warehouse management systems (iWMS).

In the past decade, applications of RFID-related techniques or systems have

increasingly been becoming popular particularly in logistics and supply chain sectors.
A latest literature review provided a summary in benefits, challenges and future trends
in RFID applications (Ming et al., 2013). Further, Sahin et al. (2007) presented a
literature review by examining the impact of inaccurate data records on inventory
management and suggested the potential of the RFID-technology to tackle this issue.

1
Wang et al (2010, 2015) investigated the trend of future generation automated
warehousing systems and proposed a framework of an RFID-enabled automated
warehousing system aiming to maximise utilisation of warehouse capacity and
efficiency of warehouse material-handling operations. Wang et al. (2010) introduced
an RFID-based warehouse management system (WMS) in which events were
managed and controlled under so-called event-condition-action (ECA) rules. Chow et
al. (2006) proposed an RFID-based resource management system (RFID-RMS) in
which a pure-integral-linear programming model using the branch-bound algorithm for
determining an optimal travel distance of a material-handling forklift in a warehouse.
Liu et al. (2006) carried out some experiments on an RFID-based resource
management system and results showed an improved utilisation of rack space and a
reduction in operational errors. Poon et al. (2009) developed an RFID-based logistics
resource management system which shares data with a warehouse database that
manages order-picking operations. Ting et al. (2012) applied an RFID-based inventory
control and management system (RICMS) into a manufacturing enterprise providing
the integrity of records in transaction and location of goods. Ross et al. (2009)
examined an RFID-based decision maker using a simulation model incorporating
major operations (receiving, storing, picking and shipping), which occur in a typical
warehouse. The decision maker can be useful for evaluating operations of a
distribution centre and examining alternative RFID implementation strategies. Xu et al.
(2013) introduced an optimisation method for implementing an RFID-enabled
warehouse management at varying SKU levels.

Integration of RFID-based systems is an important task for future generation

automated warehousing systems. Liu et al. (2009) conducted a case study by
integrating an RFID system into an enterprise resource planning (ERP) system
focusing on two modules: an electronic receiving module and an inventory transaction
module. Wang et al. (2014) discussed a number of key challenges in integration of
wireless sensor networks (WSN) with RFID systems providing an infrastructure for
data acquisition, distribution and processing in a manufacturing environment. Zhou et
al. (2008) proposed an adaptive protocol for an RFID-WSN which integrates the RFID-
based warehouse management system for tracking goods. Jehng et al. (2008)
integrated the RFID system into an automatic conveyor system in which material flow
was monitored and traced in a real time manner as each product was attached with

2
an RFID tag. Zhai et al. (2016) presented a real-time locating system (RTLS) based
on Internet of Things (IoT) seeking more accurate positions in interested sites and
scalable system topology with a flexible system capacity. In summary, it has been
widely accepted that applications of RFID techniques can facilitate automation of
storage and retrieval operations of a warehouse. Compared to a conventional
warehouse using the barcode approach, implementation of RFID systems has
demonstrated a significant improvement in warehouse data handling efficiency and
space utilisation (Liu et al. 2009). Within an RFID-based automated storage and
retrieval mechanism, each item in a tote (or a tote containing identical items) is
attached with an RFID tag so that these items can be traced, sorted and inventoried
in a real-time manner under an integrated RFID-inventory management system.
Ideally, this system can also interact with the control system of the automated storage
and retrieval mechanism. By implementing the RFID-enabled mechanism of the
proposed warehousing system, each item can also be stored and dispatched in a
storage rack (S/R) at any random location wherever a place is available for incoming
or outgoing goods.

In other relevant developments, the multi-objective optimisation approach can be

useful for obtaining a trade-off solution among conflicting objectives in logistics and
supply chain management relating to such as facility location-allocation and costs
(Gen and Cheng, 1997; Deb, 2001). Messac (2015) defined the multi-objective
optimisation as “a methodical approach to solving problems involving several
competing design objectives simultaneously”. The fundamental principle of this is that
a compromised solution can always be achievable to prioritised objectives,
accordingly. Ma et al. (2015) formulated an automated warehouse as a constrained
multi-objective model aimed at minimising the scheduling quality effect and the travel
distance. Huang et al. (2015) proposed a nonlinear mixed integer program under
probabilistic constraints for site selection and space determination of a warehouse by
minimising the total cost in inbound and outbound transportation and the total cost in
warehouse operation in a two-stage network. Lerher et al. (2013) investigated the
design in optimisation of an automated storage and retrieval system aiming to
minimise the initial investment and annual operating cost of the system. A genetic
algorithm was used for the optimisation process of decision variables. Furthermore,
Lerher et al. (2013) proposed a mono-objective optimisation approach for seeking the

3
cost-effective design of an automated warehouse. In brief, the literature review
indicated that there were very limited studies according to previous publications in
optimising the design of RFID-based automated warehousing system using the multi-
objective method.

This thesis presents a study on design theories and an integrated optimisation

methodology as part of the on-going research work for future generation automated
warehousing systems (Wang et al., 2010). The study provides a framework in
development of an integrated optimisation algorithm that is embedded into an RFID-
based inventory management system. Such a system has a capability to seek an
optimal solution to select an in-store item with an assigned priority in an order to be
pushed on and transported through an automated conveyor system to a specified
collection point within the proposed RFID-based automated warehousing system,
which has the potential capability of interacting directly with the RFID-based inventory
management system. Furthermore, this thesis also presents a study using the multi-
objective method as an aid for optimising the design of the proposed RFID-enabled
automated warehousing system. The aims and objectives of this research work are
outlined below.

1.2 Research aim and objectives

The PhD research work was aimed at investigating the latest developments of RFID-
based inventory management and systems integration for future generation
automated warehousing systems at increasingly centralised distribution centres due
to the trend of online shopping activities. Based on these, the research work was
aimed at developing an RFID-based efficient item-selection algorithm that can be
embedded and integrated into an RFID-based management system for a proposed
RFID-enabled automated storage and retrieval mechanism. This includes an
establishment of a pilot test to examine the feasibility and applicability of the developed
methodology for the system integration of hardware and software. A multi-objective
optimisation model using the multi-criterion fuzzy programming approach was also
developed as an aid of decision making to maximise the warehouse capacity utilisation
and service level of satisfying all demands of dispatching products from the
warehouse, minimise the travel distance of products from a storage rack to a collection
point and the total cost required for implementing RFIDs into the proposed automated

4
warehousing system. The developed model also supports design decisions in
determining an optimum number of storage racks and collection points that need be
established for the proposed warehouse. The detailed objectives of this research work
are as follow:

 To carry out an industrial survey relating to the issues of RFID-based management

systems for future generation automated warehouses.
 To develop a framework in design theories and/or a methodology for an RFID-
based inventory management system that can be integrated into the control system
of a proposed future generation RFDI-enabled automated warehousing system
(Wang et al., 2010).

 To create an optimisation algorithm embedded into the RFID-based inventory

management system to determine and generate a shortest travel time of selected
items in a pick-up sequence (if applicable) and route for material-handling
equipment.
 To identify and propose a cost-effective infrastructure of the RFID-enabled
communication system for the proposed automated storage and retrieval racks
(AS/RR).
 To develop an integrated program containing a scheduling a job priority algorithm
for the RFID-based inventory management system.
 To develop GUI using MATLAB allowing a visualization in changing locations of any
possibly selected item in the warehousing system in order to determine the optimal
selection of the specified/ordered item.
 To develop an optimisation mathematical model using the multi-objective approach
for 1) obtaining a trade-off solution among four conflicting objectives; 2) supporting
design decisions with the optimum number of racks and collection points that should
be established.
 To set up and execute a pilot test for examining the feasibility and applicability of
the RFID-based inventory management system for the proposed AS/RR
mechanism using the developed methodology and the integration approach.
 To carry out a cost analysis on existing inventory tracking systems using barcode
and RFID, respectively.

5
 1.3 Thesis organisation

Research background
This chapter presents a research background and addresses the research
problems which motivated this research work. The aim and objectives are included.

Industrial survey
Collected information through an industrial survey.

Literature review
This chapter provides a relevant literature review related to this work. It includes a
study in conventional and modern automated warehousing system operations.

Key components of the RIFD system

Database, RFID (Tags, reader, antenna), Connectivity of RFID system, Cables of
the RFID systems, RFID Controllers and RFID Software.

Mechanism of the proposed RFID-based AS/RS

This chapter describes Mechanism of the proposed RFID-based automated
warehouse system and its measurement.

Programme integration approach

This chapter describe the RFID-enabled warehouse management system and
programme integration approach. It also includes the RFID-related components.

Pilot study
This chapter shows the experimental results through a pilot test using the
developed optimisation algorithm and integration approaches based on the
proposed RFID-based inventory management system.

The multi-objective optimisation

This chapter presents a study in developing a multi-objective optimisation model
which was used for obtaining a trade-off decision for the proposed RFID-enabled
warehousing system design towards four conflicted objectives.

Cost analysis
A real case study in a financial analysis of the library at the University of
Portsmouth.

Conclusion and future work

The final chapter gives a summary of the research work and recommendations
of future work and enhancement.

6
Chapter 2 Industrial survey and empirical analysis

A survey was carried out to a designated group of members by answering some

questions relating to the issues of RFID-based management systems for future
generation automated warehouses based on their own experience or knowledge in
this field. The aim of the survey was to reveal the current state of warehouses in
various forms, operations of distribution centre (DC) and management and future
improvements by implementing latest developments of IT techniques and so on. The
online questionnaire sheet in details can be found in Appendix A). These
warehouses/distribution centres are conventional in which material movement is
manually performed or relies on equipment operated by human operators. Manual
material handling equipment are mostly operated within these warehouses. The size
of warehouses varies which are between 25,000 and 50,000 square feet. A
questionnaire was developed and sent via email to UK and oversees companies.
Table 1 shows a total of 33 qualified responses which were received from managers
in the relevance to their company’s warehouse management system operations. The
design of questionnaires was aimed at identifying warehouses’ current system
operations, these include the warehouse management system (WMS), information
stored within the WMS, operators’ productivity and performance, type of warehouse,
real time inventory, human errors, e-commerce services, cost reductions and rated
responses for future generation automated warehouse system.

Table 1: Qualified responses received from companies.

Valid Missing Total
Number of respondents Percentage N Percentage N Percentage
30 90.9% 3 9.1% 33 100.0%

The statistical package for the social sciences (SPSS), which is a computer program
for statistical analysis, was used to run factor analysis and correlation analysis in this
study (see Appendix A). The data of the survey in descriptive statistics were collected
through a developed questionnaire sheet that is explained further in the following
subsections.

7
2.1 Applications of database

The result of the survey shows more than 60% of respondents used the structural
database, while approximately 18% of others used the unstructured database in the
form of flat file. Microsoft is a leading provider of structural database software, almost
30% of respondents used Microsoft SQL database and Microsoft access almost 15%.
In addition, approximately 20% of respondents used oracle database and almost 18%
of respondents used MySQL database. The remaining respondents used the
unstructured database management software. Based on the survey, it clearly shows
that almost 80% of the respondents used computer database for their organisation, in
customer data and inventory tracking information for goods.

The top two primary reasons of using database were the storage of information of
customers’ details (27%) and product details (26%). Others include storage of
information of supplier details (approximately 21%) and sales data (approximately
18%). The survey shows that 50% of the respondents have their database linked with
enterprise resource planning (ERP). 44% of the respondents utilised MRP (material
requirement planning), data are extracted on a periodic basis from MRP systems. Data
from these systems can be cleaned, validated, formatted, analysed and linked with
data from many other sources. The survey also shows that 60%of the respondents
used a cloud database. In addition, slightly more than 50% of the respondents used
customer relationship management (CRM) and approximately the same amount used
application programming interface (API).

2.2 Warehouse operations

79% of respondents indicated the presence of a warehouse in their organisation. In

terms of the preference for operators' performance measurements, 30% of
respondents indicated that three main factors had an equal response: operator
accuracy, operator speed and operator efficiency; 50% of the respondents used key
performance indicators (KPIs) to measure performance within the warehouse. As
shown in Figure 1, although 60% are aware of automated warehousing systems, only
21% of the respondents had fully automated warehousing systems, a third of the
respondents had semi-automated warehousing system, and the rest of the
respondents manually operated warehouses.

8
 Figure 1: Share of manual, semi-automated and fully automated warehousing
systems.

2.3 Inventory management

The survey found that nearly 63% of the respondents used the real-time inventory. As
shown in Figure 2, 50% indicated that respondents used barcodes. 41% had voice
picking and the remaining 9% relied on other technology in the warehouse.
Predominantly, it was found that 70% of the respondents were aware of radio
frequency identification (RFID). In contrast, only 44% knew of an RFID-based
warehouse management system. Through the survey, accuracy, efficiency, operator
speed and stock loss were rated equally important. Interestingly, the survey also
revealed the estimated percentage in human errors to pick up 100 items in a manual
warehouse was 34%.

9
 Figure 2: Barcodes, voice picking or other technologies within the warehouse.

2.4 E-commerce services

According to the survey, 44% of the respondents used e-commerce services. In terms
of cost reduction to incentivise the use of e-commerce, 47% of the respondents
indicated this was the case.

2.5 Delivery

According to the survey, three main types of delivery scored quite high. The next day
delivery is most commonly used with a score of 40%. The same day delivery was the
second most popular option at 31%, followed by three days’ delivery at 22%.
Additionally, 60% of the respondents indicated that their customers require flexibility
for deliveries, although around 88% of the respondents considered the speed of
delivery as an important factor. Furthermore, it was found that about 59% of the
respondents had reserve and collect service.

Figure 3 illustrates the response rate of future generation automated warehouse

system by the respondents through the survey. It indicates that almost 79% of the
respondents scored high for the desire of future generation automated warehouse
systems as it can create a more efficient use of storage space, operate instantly and
reduce errors of manpower.

10
Figure 3: Response rate of future generation automated warehouse system.

11
Summary

A survey was conducted as part of this research work aimed to identify the current
state of warehouse operations, the type of application database used, integration with
other software (i.e., ERP, MRP…..), API usability, performance of warehouses used
type of warehouse used, real time inventory, e-commerce services and type of
delivery.

12
Chapter 3 Literature review

Warehouses primarily serve as storage points in a supply chain network between

origins of raw materials and destinations for consumption of final products (Wang et
al., 2011). Figure 4 illustrates a typical warehouse in operations, functional areas and
product flows (Tompkins et al., 2003). Warehousing operations can be divided further
into the following functions (de Koster, 2007; Lambert, 1998):

 Receiving: This consists of unloading the products from transportation vehicles

to receiving docks, inspection of products for decencies or missing products,
and updating warehouse inventory records to react changes.
 Transfer or put away: which includes operations in moving products from
receiving docks to assigned storage locations, shipping docks or other areas in
the warehouse, and moving products between these areas,
 Order picking: which consists of collecting required quantities of specified
products from storage locations to satisfy customer orders,
 Shipping: which includes operations in loading products onto transportation
vehicles, inspection of products to be shipped, and updating warehouse
inventory records, sorting and packing products.

Figure 4: Typical warehouse functions (Tompkins et al., 2003).

These operations may be performed under a warehouse management system (WMS)

which controls the storage and movement of materials and processes the transactions

13
within a warehouse. The successful warehouse resources management is one of the
most vital factors in terms of handling warehouse operations effectively and efficiently
leading to satisfaction of demands of suppliers and customers (Poon et al., 2008).
Inventory management is a key in logistics and supply chain management because
inventory decisions are often a starting point for other business and operation
activities. In addition, different organizations of warehouses may have different
inventory management objectives (Marphy & Wood, 2011).

3.1 Inventory management systems

3.1.1 MRP/ERP

In other aspects, to develop cost-effective and efficient operations in manufacturing,

reduction of costs and optimisation of operations is one area that needs to be
addressed. A large part of this task relies on how manufacturing companies plan their
operations, control inventory and organise logistics. Material requirements planning
(MRP) is a time phased priority-planning technique that calculates material
requirements and schedules supplies to meet demands across all products and parts
in one or more plants (Hamid et al., 1991). MRP determines the quantity and timing in
acquisition and demand of dependent items that needs to satisfy master schedule
requirements. One of its main objectives is to keep the due date that equal to the need
date aiming to eliminate material shortages and excess stocks. The MRP system
breaks a component into parts or subassemblies, and generates plans for those parts
to arrive into stocks when needed (Ghobbar et al., 2004). In the past decades,
information technology has played a major role in designing and implementing MRP
systems as it provides a platform for information integration from manufacturing needs
that link with customer demand and existing inventory levels. MRP techniques are
developed and used for bills of material (BOM) and net material requirement plans for
future production. MRP are computer-based planning and scheduling systems
designed to improve and enhance the management control of manufacturing.
Extension of MRP systems may have capability to capture other manufacturing related
requirements.

Several studies explored the effectiveness of using MRP in manufacturing sectors.

Landvater et al. (1985) carried out a survey of 1800 companies and the result showed
that 60% of these companies used MRP/MRPII. Advantages reported by the

14
companies are: (1) reduction in inventory levels, (2) improvement of consumer
services, (3) increase in productivity and (4) reduction in raw material costs. other
management oriented inventory control techniques such as Just-In-Time (JIT) and
Enterprise resource planning (ERP) also plays an important role (Lwiki et al., 2013),
although small and medium enterprises and some larger companies still rely on past
data and experience to determine amount of materials to purchase. ERP is an
information system that supports company management integration and or an
interaction among a group of cooperating enterprises ERP may also be as marital
planning system or more commonly known as manufacturing resource planning. In the
past decades, ERP has been widely used for managing the following business
activities:

 Financials.
 Order processing, logistics.
 Sales, marketing, customer relationship management.
 Manufacturing processes & quality control.
 Service management.
 Budgeting & general forecasting.
 Human resources.
 Payroll.
In supply chain sectors, most ERP systems are used to collect, manage, store, and
covert data from different business units across the organisation. Yüksel et al. (2011)
introduced middleware software (ERP/MRP) for business models using RFID
technology in a supply chain management. Liu et al. (2009) combined an RFID
technology with an enterprise resource planning (ERP) system for resolving
various wafer receiving and inventory transaction problems.

3.1.2 Warehouse management systems (WMS)

WMS is an inventory tracking software-based system in which it processes the

incoming/outgoing transactions, including shipping, receiving, stock allocation, order
picking and fulfilment. A WMS monitors the warehouse progress of items as follows:

 Inventory location assignments.

 Order picking & fulfilment management.
 Warehouse capacity management.

15
  Capability for data management.
 Cross docking management.
 Put-away and picking optimisation.
 Labour utilisation.

3.1.3 Lean management

Lean management is a concept that was first applied in the Toyota motor corporation
in the 1950 (Hines et al., 2004). It was originally developed and applied by the
Japanese due to the lack of natural resources and intense domestic competition in
automobile industry in Japan. This method also made it possible for Toyota and other
firms to achieve high levels of quality and productivity in manufacturing (Cusumano,
1994). Lean manufacturing helps in enhancing production processes and boosting up
the employees’ job satisfaction (Singh et al., 2010c). Gupta et al. (2013) argued that
the traditional methods in dealing with manufacturing activities rely on a good
management of inventory of the system, whereas lean manufacturing, however,
considers the inventory as a waste in production and lean manufacturing focuses on
the elimination of wastes in production, i.e., the waste of overproduction, the waste of
waiting for parts to arrive, the waste of conveyance, the waste in processing, the waste
of inventory, the waste of motion and the waste of rework (Ballé et al., 2005). Figure
5 illustrates the seven wastes identified in Lean Manufacturing.
By implementing lean manufacturing properly, it may increase efficiency in production
outputs and operators and decrease in process inventory (or work in progress) levels
of ongoing and finished goods (Seth et al., 2015). Ultimately, lean manufacturing
system aims at eliminating all wastes if possible within the entire organisation. Dennis
(2007) stated that the main goal of a lean manufacturing system is to produce products
of higher quality at the lowest possible cost in the least time. Chen et al. (2013) applied
lean production and radio frequency identification (RFID) technologies to improve the
efficiency and effectiveness of supply chain management.

16
 Over
Defects production

Inventory

Waiting 7
Wastes

Over

Transportation processing

Motion

Figure 5: Seven wastes in lean manufacturing.

3.2 Manual warehouse

In manually operating warehouses, operators travel to a pick location, pick goods, and
move these goods to the delivery point. Operators may also engage with forklift and
conveyors that transport these goods in certain locations in the warehouse. This type
of manual warehouse may lead to (Wang et al., 2010):

 Material handling inefficiency.

 Poor space utilisation.
 Slow process for stacking and dispatching.
 Low variety and high level of SKUs.
 Frequent human errors.
 Costs of labours Training.
 Accidents.
 Outdated inventory data.
 Poor inventory visibility.
Thus, many companies intend to replace traditional warehouse with automation.

17
3.3 Automated warehouses

Modern warehousing systems can be classified as mechanized (conventional) or

automated. A combination of labour and handling equipment is utilised in mechanized
warehousing systems to facilitate receiving, processing, and/or shipping operations.
Generally, labour constitutes a high percentage of the overall costs in mechanized
systems. Automated systems on the contrary, attempt to minimise the labour element
as much as possible by flaking capital investment in equipment. An automated system
operates faster and more accurate than a mechanized system. During the last two
decades, researchers have developed a number of solution procedures in warehouse
design and operation (Ashayeri et al., 1985).

Conventional automated warehouses use extensive conveyors, sortation equipment,

automated storage and retrieval systems (AS/RS), and other material handling
solutions. Qiu et al. (2013) proposed the concept of a supply hub addressing the
construction of warehouses for manufacturing enterprises located within an industrial
park. The research focused on evaluating the value of freight consolidation using the
SHIP (Supply Hub in Industrial Park) approach. The warehouse is used as a switching
facility rather than as a long-term storage house. Attention is paid to higher inventory
turnover, lower operating cost, and shorter cycle time. Simulation methods are also
widely used for automated warehouse design for seeking maximum space utilisation
and increasing material flow to achieve a higher productivity and cost reduction that
fits into the objectives of the effective supply chains. Advantages and disadvantages
using automated systems, however, can be held against each other. Grant et al.
(2006) listed some of the advantages and disadvantages as follows:

Advantages

 Cost reduction in labour.

 Increase of output rate.
 Reliability and time of service will be improved.
 The amount of material handling will be reduced.
 Increased accuracy level.
 Increased availability.
 Better working/ergonomics condition.

18
Disadvantages

 Initial capital cost might be high.

 Risk of downtime and unreliability of equipment.
 Software related problems.
 Lack of flexibility to respond to changing environments.
 Problems with the capacity.
 Maintenance interruptions and the related costs.
 User interface and training.
 Work acceptance.
 Obsolescence.
Automated warehouses are suitable for both boxed and palletised goods stored on
shelves or racks. The automated storage and retrieval racks (AS/RR) are designed for
undertaking a high volume of loads being moved into and out of storage racks. The
load handling area is situated at one end of the shelves or racks and conveyor systems
are installed to mechanise the movement of items from one place to another. The
inventory of warehouse is controlled by the warehouse management system (i.e,
WMS), which records information of all the in-store goods. The benefits of these are
(Lerher et al., 2010):

 Automation of the product entry and exit operations.

 Saving space and time devoted to storage tasks.
 Elimination of manual handling errors.
 Ease of inventory data management, updating and control.
 Retrieval equipment guided by management software, which coordinates all
warehouse movement operations.
 Reduction of damage and loss of goods.
 Decrease of the number of warehouse workers.

3.4 Internet of things (IoT)

Advances in enterprise information management, system control, inventory

management, machines-to machines (M2M) communication, human-to-machines
interaction and item tracking are shifting to a new paradigm of Internet of Things (IoT)
for future industrial enterprise information systems (Zuehlke, 2010; Niu et al., 2013).
Internet of Things (IoT) is defined as the ability of various things to be connected to

19
each other through the Internet (Evangelatos et al., 2012). The goal of the Internet of
Things is to enable things to be connected anytime, anyplace, with anything and
anyone ideally using any path/network and any service (Vermesan et al., 2013).
Examples include personnel positioning (Werb, et al., 1998), tool management,
material and product tracking (Won et al., 2009; Cao et al., 2013; Zou et al., 2014),
process control (Ma et al., 2011), assembly (Rakotondrabe et al., 2011) and mobile
robot localisation (Song et al., 2012; Squire et al., 2014). The operating distance can
be on a large scale, for example, a warehouse, or on a small scale, for example, a
machine area, with required localisation accuracy from a few meters to centimetres.
Consequently, a reliable and scalable real-time locating system (RTLS) becomes
essential to enable enterprise location intelligence. In the Internet of Things (IoT)
paradigm, many of the objects in the warehouse can be on the network in one form or
another. Radio frequency identification (RFID) and sensor network technologies will
rise to meet this new challenge, in which information and communication systems are
invisibly embedded in the industry environment. This results in the generation of
enormous amounts of data which must to be stored, processed and presented in a
seamless, efficient, and easily interpretable form.

Internet of Things has some exceptional advantages as a new technology, to achieve

the objects of the intelligent identification and management, improve resource
utilisation and efficiency. Internet of Things can use RFID technology to achieve the
exchange and sharing of information is an important means to break through the
bottleneck of the times the items transparent tube through the open computer network
(Yanhui, 2013).

Cloud-based inventory systems can track items in real-time by scanning them in and
out of stock locations. Products usually have either an RFID tag or barcode label so
they can be scanned and identified by the system. Currently, this is how inventory
management systems can give an organization visibility into their inventory levels,
expiration dates, item location, forecast demand, etc. Within the IoT, the ability to track
and communicate with RFID-tagged products can be greatly increased and will be
able to communicate the information with an inventory system. In this case, RFID tags
will be built into objects, which will then be able to send information about the status
of this object. Such an ability to track, and monitor inventory can be improved through
integration of IoT. Within the IoT, it will eliminate human-to-human or human to

20
machine interaction. Nevertheless, RFID is currently a better system for the inventory
management system, this is due to the higher cost of the IoT which is still under
development, more research is still required to make the IoT a reality.

3.4.1Technology Readiness Levels

TRLs are a systematic metric/measurement system that supports assessments of the

maturity of a technology and the consistent comparison in maturity between different
types of technology. As stated previously, academic interests in RFID technology have
been growing in recent years particularly in supply chain management and logistics
functions of retail sectors. McGuinness (2008) examined the readiness of the high-
technology manufacturing industry to adopt and use RFID technology in its products
and processes through a case study analysis of the Irish medical devices sector. Ly
(2006) presented a readiness assessment method for the maturity of RFID technology
in order to implement an automated checkout system at a supermarket. The readiness
assessment results indicated that the levels 1, 2, 3 and 4 reached the maturity of the
RFID technology. For the levels 5 and 6, the technology was still in yellow state
because the current RFID tags do not transmit well on certain products such as liquids
or metals. This limits the overall benefit of RFID until the problem is solved. In addition,
the automated checkout system prototype neither demonstrated nor was completely
tested in level 7 and 8. Finally, the level 9 was not achieved because the system failed
to perform operations. Ebrahimi et al. (2015) conducted a study aimed at comparing
the readiness of Shiraz University of medical sciences hospitals for implementation of
RFID system. The study showed that the readiness level of the hospitals was
moderate and the total readiness of hospital was higher than other hospitals. Despite
the general acceptance of the superior benefits that RFID technology offers over
competing auto-identification technologies, the level of adoption of RFID technology
within manufacturing sectors remains small. Nevertheless, there appears to be a trend
amongst many organisations that it may only be a matter of time before adoption of
this technology becomes widespread (McGuinness, 2008).

3.5 RFID technology and applications

It is difficult to trace the history of RFID technology back to a well-defined starting point;
Figure 6 illustrates the evolution of RFID technology from 1940 to 2000. The important
decade for RFID technologies was in 1990s when the United States developed a

21
regionally compatible electronic toll collection system using a signal tag and signal
billing account vehicle to access the highways and bridges of several toll authorities
(Landt, 2001). RFID is a term used to describe a system that transfers the identity
information of an object or person wirelessly using radio waves. RFID is an automatic
identification method in which an FRID system can store and remotely retrieve data
using chips called RFID tags and RFID readers. A typical RFID system uses passive
RFID tags to hold information, much like barcodes, but these tags can contain other
more useful user-define information. These types of RFID tags are read by antennas
that power RFID tags by interrogating them with a radio signal. There are also long-
range RFID systems that use active and semi-active battery-powered RIFD tags rather
than using interrogating radio frequencies. RFID tags are often used to identify pallets
and large packages in logistic and supply chain sectors. Implementation of RFID
systems has also become popular in healthcare settings as it allows nurses access
medication cabinets, track medication inventory, equipment and staff, and patients
within a RFID system. In supermarkets or big stores, implementation of an RFID-
system can track and trace the stock from a supplier to a customers’ shopping trolley
(Brown, 2007).

7 Very early adopter

implementations of RFID
Early
6 explorations
of RFID Development of
technology the theory of RFID
5 Emergence of
laboratory
standards,
experiments
4 Radar refined RFID widely
and used, deployed
major becoming a
3 development part of
effort in WWII Start of applications everyday life
RFID field trials Commercial applications
2 invented of RFID enter mainstream
in 1948
1

1940 1950 1960 1970 1980 1990 2000

Figure 6: Evolution of RFID technology (author’s compilation).

RFID offers a number of advanced properties in terms of its ease of communication

and real-time information (Saygin et al., 2007). Leung et al. (2007) summarised the

22
benefits of using the RFID technology. Roberti (2003) showed a study that out-of-stock
items with RFID were replenished three times faster than items using the standard
barcode technology. Sounderpandian et al. (2007) presented a study in costs of
implementing RFID technologies in a supply chain. Due to recent developments in
data processing and microelectronics, RFID components are becoming smaller, much
less expensive and more effective. The RFID applications also includes sectors of
retail, textile, healthcare, automotive and luxury goods. For instance, in the competitive
textile and fashion industries, RFID is an effective means by which it provides a quick
response to customers in identifying counterfeits and genuine products. In
pharmaceutical industry, RFID are used for tracing the whole life cycle system in raw
material supplies, production, transportation and storage. Figure 7 shows the report in
penetration of active RFID into different application sectors over the next ten years
(2009–2019).

Figure 7: RFID applications into different sectors (Harrop et al., 2010).

23
The latest development of RFID techniques has gained even greater attention in
logistics and supply chain sectors where the implementation of a RFID-integrated
information management system is becoming increasingly desirable (Bagchi et al.,
2007). Ne´meth et al. (2006) examined the challenges and possibilities applying RFID
systems into supply chains. Compared to the barcode approach, the implementation
of RFID systems has demonstrated a significant improvement in supply chain
management through the reduction of inventory losses and efficiency of inventory
processes. Chen et al. (2010) proposed RFID-based enterprise application integration
(EAI) approach for the real-time management of dynamic manufacturing processes. A
prototype system was developed to demonstrate the applicability of the proposed
method in a shop floor environment. Brusey et al. (2009) focused on the issue of
correctly identifying, tracking and dealing with aggregated objects in customised
production using RFID. Chao et al. (2007) reviewed the trends on RFID technologies
in supply chain management. Delaunay et al. (2007) presented a survey in the causes
of inventory inaccuracies in supply chain management. Dolgui et al. (2008) also
presented a literature review on RFID technology in supply chains outlining the
advantages of RFID technologies in inventory management and analysing privacy and
authentication properties of RFID technologies. Kärkkäinen (2003) discussed the
possibility of RFID being implemented to improve supply chain efficiency for short
shelf-life products. McFarlane et al. (2003) looked into certain obstacles faced by the
processes of supply chain logistics in terms of transportation, shipping, receiving, and
in-house operations to assess and analyse the potential benefits of this technology.
Turcu et al. (2007) examined the potential benefits generated by integrating a RFID
system with business-to-business (B2B) applications and subsequently proposed an
integrated RFID-B2B system. The system offers multiple performance levels for
varying systems or application needs, and it can be readily personalised to meet users’
demands.
In manufacturing sectors, some companies used RFID technology to track
components or parts at each stage of the manufacturing status of operations that take
place during the production process. Data exchanged from this transaction can then
be stored within the ERP system, which is made accessible to managers and decision-
makers. There is vast literature on the use of RFID to better manage inventory and
improve the operations in a supply chain. Lee et al. (2004) analysed the level of
inventory visibility, accuracy and shelf replenishment policies of inventory operations.
24
Doerr et al. (2006) used the factorial structure for analysing the non-cost related
benefits and the ROI (return on investment) in RFID implementation. In recent years,
a great deal of demands have been arisen from logistics and supply chain companies
for an improved inventory visibility by recognising the importance of the RFID-based
management system that can be implemented into the automated warehouses.

3.5.1 RFID applications in warehousing systems

The growing trend for online shopping is likely to gain even greater prominence every
day. More and more customers nowadays prefer to order their goods that can be found
online and these customers also demand a faster delivery of ordered goods to be
dispatched directly to their door steps. A number of studies through a literature review
suggested that future generation warehouses may be designed and implemented as
more centralised distribution centres that partly replace traditional supplier and retailer
stores or shopping malls in the commercial street. As a result, these suppliers and
retailers require more efficient distribution centres in which a novel design of
automated storage and retrieval systems are increasingly desirable. Current
automated warehouses are equipped with expensive equipment including palletising
robots, carton flow order picking systems, automated guided vehicles (AGVs), rotary
storage cabinets, and automated storage and retrieval systems (AS/RSs). These
facilities can partially replace human workers for performing operations of picking and
placing items and vertically improve space utilisation of a warehouse. Furthermore,
implementation of IT capabilities has also demonstrated the potential for greater
improvement opportunities in terms of tighter inventory control, shorter response time
and greater variety of SKUs (stock keeping units). These IT capabilities can be
implemented by using smart-labels such as barcodes, radio frequency identification
(RFID) tags and automatic identification (Auto-ID) sensors, together with cable and/or
wireless communication networks and integrated warehouse management systems
(iWMS). A study by Wang et al. (2010) indicated that an RFID-enabled automatic
warehousing system has several advantages over the conventional automatic
warehouse as described above. In such a novel design of storage and retrieval
mechanism, an item, which is attached with a RFID tag, can be traced, sorted and
inventoried in a real-time control manner but also the item can be stacked on the
storage racks randomly at wherever a place is available for incoming goods and
distributed from anywhere throughout the system.

25
Unlike the barcode scanning method, radio frequency identification or RFID
technology uses wireless non-contact radio wave communication mechanism to
identify an object by transmitting a unique identity code that contains information of
this item, each item is attached with a passive RFID tag (containing a silicon chip) and
the captured data through the signal transmission between an RFID tag and a reader
can be uploaded to a warehouse computer database to look up information relating to
this item. Some RFID systems use active and semi-active tags which are powered
with batteries to communicate with the reader. With this approach, the travel distance
of the signal transmission between an active RFID and an RIFD reader can be much
longer than passive RFID tags. It has also been seen as a trend that RFID systems
may replace barcode systems in future. With RFID technologies, warehouse inventory
can be tracked more accurately in a simple, timely and more efficient manner leading
to a significant reduction of data processing time and labour cost, compared to the
barcode method as each barcode has to be scanned individually, which is a labour
intensive and time-consuming process. An RFID tag can also store much more user-
predefined information than traditional barcodes. Figure 8 shows the advantages and
disadvantages of using barcodes and RFID methods, respectively (McCathie, 2004).
The cost of RFID tags is one of the main constraints for the full RFID implementation
within supply chains (Chomka, 2003).

26
 RFID

Advantages
Barcodes

Disadvantages

Figure 8: Barcodes vs RFID (McCathie, 2004).

27
The cost of RFID tags depends on their type. Reported by RFID Systems in the
Manufacturing Supply Chain (2003), although there is a 5–35% decrease of labour
costs using RFID tags, prices of active or semi-passive RFID tags (at least $1 per tag)
are still a hindrance allowing the economic application; the applications were mainly
suitable for high-value goods over long ranges. By comparison, barcodes are the least
expensive way offering a simple method to capture the data. Barcode labels cost less
than 2 cents per label while RFID tags are at least three times more expensive per tag
(Williams et al. 2014). The precise cost of RFID tags varies depending on the
underlying RFID technology and the number of items attached by RFID tags. Typically,
passive RFID tags are between 7 and 20 cents. Table 2 shows a summary in
performance measures between barcodes and RFID tags through a literature review.

Table 2: Summary of a comparison between barcodes and RFID tags (Source:

Çakıcı et al., 2009 and Hedgepeth, 2009).

Criteria Barcodes RFID

Convenience Require line of sight scan Automatic scan without line of
sight
Efficiency Cannot support batch reading Can read multiple tags at once
Accuracy Susceptible to misreads and Reduced human error and
human error misreads improves data
accuracy
Traceability Limited traceability. Some harsh Allows detailed tracking and
manufacturing processes (link tracing of process status,
baking) make barcode tacking input/output, and the time that
impossible each processing step was
performed.
Speed Process information not in real- Real-time process information
time
Reliability Barcodes are easily dirtied or
RFID tag can survive harsh
scraped in harsh manufacturing
environment such as high
environment temperature
Automation Need more human labour to Replace or reduce human
collect and track process data
labour in data collection and
limited process information tracking
Information Limited process information Vast amount of detailed process
information
Storage (data) Allows only centralized data RFID tags carry large data
storage capabilities

28
Table 3 shows various RFID applications in different sectors with expected outcomes
based on a literature review.

Table 3: A number of case studies in RFID applications through a literature review.

Authors Industry Applications Results

Warehouse Four inter- RFID technologies are useful to apply
Lefebvre et of a specific related firms because it can improve the existing
al. 2006 supply chain from three processes, provide a new business
echelons of the model and increase the communication
supply chain. between supply chain actors
RFID-EPC network can enhance the
Mobile Four different operational processes such as shipping,
Wamba et al. business to companies. receiving and put-away processes and
2007 business e- business processes through automated
commerce activities.
RFID employment can significantly
change processes and human
Tzeng et al. Health care Five hospitals in resources of the organisations, enhance
2008 industry Taiwan customer satisfaction and improve
efficiency and flexibility of process
redesign. They note that re-engineering
application optimises systems.
RFID technology can significantly
Construction High-tech improve supply chain control and
Wang et al. supply chain factory construction project management by
2007 environment building in improving the efficiency of operations
Taiwan and also by providing a dynamic control.
Eight members Proposed different models with varying
Hou et al. Printing of the complexity and provide quantitative cost
2006 industry Taiwanese and benefit analyses of RFID
printing industry. technologies integration.
Intelligent Three Demonstrated the technical feasibility of
Ergen et al. components experiments in RFID technology in supply chains, and
2007 in three types of indicated that active UHF RFID
engineered- components. technology can create intelligent
to-order components efficiently.
(ETO)
management
Huber et al. Retail supply Nine Australian RFID can minimise losses in the supply
2007 chain RFID vendors chain and the visibility of stocks is the
and main shrinkage factor that RFID can
associations improve
Three-level Cooperation between the business
Baars et al. Retail supply supply chain – intelligence and RFID technologies
2008 chain Chinese enhances supply chain operations, but a
manufacturers, cost-benefit analysis should be realised
a consolidator

29
 and a GDC in
Germany
Fast-moving Three-echelon Both in the integrated and non-
Bottani et al. consumer supply chain integrated scenarios, RFID technologies
2008 goods which contains at pallet level can provide benefits for all
(FMCG) manufacturers, echelons however manufacturers
supply chain distributors and cannot obtain positive revenues
retailers of because of the high cost of case level
FMCG. tagging.
Automotive RFID significantly reduces the order to
Mourtzis et industry delivery time so that customisation
al. 2008 orders can be realised in spite of market
variation.
Zhang et al. Automotive Showed the benefits of RFID technology
2010 industry in the green supply chain (GSC) area
Hardware and software applications
Kim et al. Retail supply Numerous U.S. influence RFID benefits in inventory
2008 chain and Korean management for U.S. retailers while, for
retailers, Korean retailers, it can improve the
efficiency of store operation and
demand management.
Kim et al. Retail supply 278 adopting Showed the organisational needs,
2010 chain organisations perceived factors and organisational
readiness.
Three main objectives: simplifying RFID
Poon et al. Warehouse integration, improving the visibility of
2009 warehouse activities and the
performance of the warehouse.
RFID technology integration improves
Pigni et al. the system business process and
2008 Fashion provides an inter-organisational
industry information system that improves the
efficiency and effectiveness of the entire
supply chain

30
3.6 RFID-based automated warehouse systems

A modern warehouse combines the state-of-the-art mechanical hardware with

information systems to achieve warehousing operations for a high variety of loads.
The key part of automated warehouses is automated storage and retrieval systems
(AS/RS) (Lerher et al., 2010). The AS/RS is a material handling mechanism which is
a key element of automated warehouses or distribution centre. The basic components
of AS/RS are storage racks (SR), automated storage and retrieval (S/R) machines,
and accumulating conveyors. AS/RS offers the advantages of fewer material handlers,
better material control and more efficient use of storage space. The disadvantage is
that the cost to maintain such equipment is relatively high once it is installed. Figure 9
illustrates the overall design of an RFID-enabled automated warehouse model by
Wang et al. (2010). It consists of an individual module in which all levels of output
conveyors are interconnected to equal levels of a loop spiral conveyors to collection
points. These conveyors, which also act as delivery paths as well as buffers, form the
output zone of the warehousing system. The module is designed as standardised
element for manufacturing and assembly, although each module can be of different
sizes and arrays in a module can be configured easily in many different ways, i.e.,
capacity of warehouse is adjustable. In the warehouse, items are pre-loaded onto
pallets named totes; totes need not be of identical sizes. Each item in a tote (or a tote
containing identical items) is attached with an RFID tag and each item in a tote is
thereafter tracked and manipulated by the developed RFID-inventory management
system throughout the warehouse.

31
 Figure 9: The overall layout of the 3D warehousing system (Wang et al., 2010).

As the travel time covers a substantial part of picking processes in warehouses, an

appropriate optimisation in sequencing of picking lines within a batch is crucial to
achieve high efficiency. Many researchers have focused on minimisation of travel time
for an item travel to a collection point using various methodologies (Shaw, 2002).
Lerher et al. (2006, 2010) presented an analytical travel time model for multi-aisle
AS/RS considering parameters in acceleration, declaration, deceleration and
maximum velocity. Yahong et al. (2003) introduced a new type of storage/retrieval
(S/R) mechanism to handle loads and proposed a travel time model under the dwell
point policy. The dwell point in an AS/RS is the position where the S/R machine resides
when the system is idle. The dwell point is selected such that the expected travel time
to the position of the first transaction after the idle period is minimised. An effective
dwell point strategy may reduce the response times of the AS/RS, since the S/R
machine typically performs a sequence of operations following an idle period (Vasili et
al., 2006). Similarly, Hu et al. (2005) developed a new type of S/R mechanism and
proposed a travel time model under the stay-dwell point policy, i.e. the platform
remains where they are after completing a storage/retrieval operation. Liu et al. (2014)
carried out a research on a travel time model of a new compact storage system which

32
is an extension of conventional automated storage and retrieval system (AS/RS).
Agarwal et al. (2015) proposed an analytical model for the computation of travel time
for automated warehouses with the aisle transferring S/R machine. Lerher et al. (2015)
developed an analytical travel time model for the computation of travel (cycle) time for
shuttle-based storage and retrieval systems (SBS/RS). The proposed model
incorporates the operating characteristics of the elevators lifting table and the shuttle
carrier with parameters in acceleration and deceleration and the maximum velocity.
The study conducted by Schenone at al. (2016) aimed at providing a new methodology
suitable for computing the travel time for an AS/RS system based on the class-based
storage assignment policy.

Early warehouses gave little consideration to space efficiency, order picking methods
or material handling. Today, tighter inventory control, shorter response time and
greater variety of SKUs (stock keeping units) are most important challenges for
designing modern warehouses. Although these challenges can be prompted in part by
adoption of a number of management philosophies such as just in time or ‘lean’
methods as the warehouse inventory policy, the latest advanced IT and
communication technologies play a vital role in bringing effectiveness and efficiency
to the warehousing system through speedy information processing and internal and
external connectivity. However, the degree of usage of the latest IT tools such as
Radio Frequency Identification (RFID) and communication systems depends on
management policy, product type, market competition and customer expectation.
Some leading e-commerce retailers have used automated storage systems to gain a
greater efficiency in progressing their e-tail operations. As the materials handling
revolution continues, there will be more techniques applied in design and operation of
production and distribution systems. For instance, routing sequence of older picking
can be programmed into the WMS (warehouse management system) (Moeller, 2011).
Order picking can be defined as the activity order requested by a customer. Order
picking can be performed in a variety of high-level approaches under an order picking
system (OPS). In a manual operating warehouse, picking processes are seen as the
most labour-intensive and costly activity and the cost of order picking is estimated to
be as much as 55% of the total warehouse operating expense (De Koster et al., 2007).
Poor performance in order picking can lead to unsatisfactory service level and high
operational cost for warehouses. Figure 10 shows operating costs (e.g., shipping,

33
picking, storage and receiving) in a typical warehouse (Tompkins et al., 2003). As
shown in the Figure, the cost of order-picking operation dominates up 55% of the total
operating cost. Given this economic importance, the decision supports for planning
and controlling the order picking processes within a WMS is a central task.

Figure 10: Operating costs in a typical warehouse (Tompkins et al., 2010).

Intelligent control software is required to realise the potential offered by the

warehousing system design shown in Figure 9. Input variables in the control software
can be classified into three categories, namely application parameters, hardware
parameters and software parameters. Application parameters may include such as
numbers of different SKUs handled by the system, the storage information of each
SKU, the number of items per request, the number of requests per time unit, and the
number of concurrent users etc. Hardware-related parameters may include factors
such as the height and depth of each rack, the number of racks per aisle etc. Software
parameters may cover scheduling and optimisation algorithms used for job priorities
to assign or dispatch items to/from arrays of each individual storage modular element
to minimise the travel time and minimise the operation cost as shown in Figure 6. In
order to achieve this, a material-handling solution should be provided based on optimal
algorithms in order to generate the shortest pick-up sequence and route for the
material handling equipment.

34
Summary

This chapter presents a literature review study in which it examines developments of

existing manual and automated warehouses, and warehouse inventory management
systems and methods. The chapter also examines RFID and sensor network
technologies including the background of the RFID and its application mainly for
supply chain and logistics sectors and review the applications that have been
established by researchers. Besides, the RFID- based warehousing system is
illustrated as the basis of this research work. A comparison between barcodes and
RFID methods was provided within this chapter. A number of case studies in RFID
applications were included. The chapter illustrates a typical RFID-based automated
warehousing system of which inventory of incoming and outgoing items in a
warehouse is controlled under the warehouse inventory management. Travel time
covers a substantial part of picking processes in warehouses in which the item can be
dispatched to travel from the storage rack to the collection point. The relationship
between the operation and cost in the warehouse was also explained. In conclusion,
through a comprehensive literature review, there was no study which was found in
reporting design theories and integration approaches for the RFID-based future
generation automated warehousing systems. By leveraging the latest supply chain
technology and the Internet of Things (IoT), a smart warehouse can now serve as a
hub to boost efficiency and speed throughout the entire supply chain, internet-enabled
devices and technology may profoundly change logistics management in the near
future.

35
Chapter 4 The RFID-enabled inventory management system

Figure 11 shows the key components of the RFID-enabled inventory management

system used for this study, which are described in details as follows:

Figure 11: The key components of the RFID-enabled inventory management system.

4.1 RFID tags

RFID data are store in an RFID-based warehouse database as shown in Figure 9. The
supporting software for the RFID-based database can be developed using Oracle,
SQL Server, Postgres, or MySQL. Figure 12 shows the key components of a typical
RFID chip. A typical RFID tag contains a small chip, also called an IC, and RFID tags
can be attached to everything that needs to be tracked. The circuit connects with an
antenna, which can be integrated with a vast array of labels, security tags, and apparel
hang tags, and other kinds of industrial asset tags. The chip of each RFID tag contains
a pre-programmed serial number or unique identifier (UID) assigned along with
memory, which includes a unique tracking identifier that stores the product’s unique
electronic product code (EPC) and other data on memory contained in its chip, which
enables it to be read and tracked using RFID technology (Brown, 2007).

36
 Modulator Encoder Logic

Antenna
AC/DC
convertor
Antenna Power
control
Memory
Decoder

Figure 12: key components of a typical RFID chip.

As shown in Figure 13, different types of RFID tags are available in the market to suit
usages of tagging not only products but also people (Gambon, 2009). For instance,
some of the features of these RFID systems allow nurses to access medication
cabinets, and to track medication inventory, equipment, and staff as well as patients
by substituting wristbands and charts with a combined patient chart and medical
bracelet with an implemented RFID system (Lingle, 2007; Park, 2014). Another
application of RFID systems is in supermarkets and large stores, whereby miniscule
RFID tags are assigned to every item on every shelf, so that the supermarket is able
to track and trace stock from the factory from their distribution chains to customers’
trolleys (Roberti, 2004).

Figure 13: A variety of RFID tags (Impinj Company).

There are several types of RFID tags depending on the purpose of their applications.
The RFID tag can be adapted to suit various design requirements according to the
types of items to which the tags will be attached (e.g., wrist belt, buttons, jewellery or
even a tattoo). The amount of information each RFID tag can carry vary depending on
the capacity of its computer chip memory. The type of data carried by each RFID tag
also varies. Jian (2010) lists three types of RFID tags according to the specific
requirements of application. These include passive RFID tag, active RFID tag, and
semi-active RFID tag. RFID tags are designed with multiple antennas, making them
37
more reliable than RFID tags with only one RFID antenna as those RFID tags that
have a single antenna can be affected by “dead zones”, where parts of the RFID tag
may remain unresponsive to incoming signals resulting in the RFID tag not acquiring
sufficient energy for powering its internal chip and inlay therefore preventing
communication with the reader (Kodaka et al., 2009; Taguchi & Mizuno, 2006). RFID
tags with dual antennas can bypass this obstacle, increasing their readability, provided
it matches the chip design (Lee & Yu, 2015).

4.1.1 Passive RFID tags

Passive RFID tags do not have access to their own source of power; instead, these
RFID tags are entirely reliant on the RFID readers to activate them. When a passive
RFID tag is triggered, it transmits the information that is stored in the RFID tag through
its antenna to the RFID reader antenna (Hunt et al., 2007). The RFID reader analyses
the information it receives and acts according to the requisition of the RFID system.
Passive RFID tags are typically used cards to authenticate valid users (Jian, 2010).
As passive RFID tags can connect with the RFID reader using both inductive and
propagation coupling due to having an on-board RF transmitter. In addition,
information cannot be transmitted without the RFID reader being in the same vicinity.
Also, the communication is one-way only, in which an inquiry is made by the RFID
reader first, followed by the response from the RFID tags.

The frequency for passive tags which commonly function well are 860–960 MHz in
UHF (ultra-high frequency), 13.56 MHz in HF (high frequency) (Liu et al., 2014; Park
& Yu, 2008), and 124–125 kHz and 135 kHz in LF (low frequency) band (Park & Yu,
2008; Yang, 2010). There are also different frequency bands that RFID can legally use
depending on what country the RFID is being implemented in; it must therefore be
legally compliant to the frequency regulations of the country in which it will operate
(Brown, 2007; Yang, 2010). Moreover, it is often thought that RFID can provide
constant and continuous location data (Ferguson, 2007).

4.1.2 Active RFID tags

An active RFID tag, unlike a passive tag, has its own battery power, allowing it to
actively broadcast the data stored on it via an antenna, even without the presence of
a RFID reader. An active RFID tag is often used for sensing pressure, humidity,
temperature, etc. Typical applications for active RFIDs are wireless sensors. In

38
comparison, semi-active RFID tags are also battery-powered; however, this type of
RFID tag is able to turn itself off after being triggered by a given action from an RFID
to transmit data to the RFID reader (Jian, 2010). Active RFID tags typically function in
frequency range of 455 MHz, 2.45 GHz, or 5.8 GHz, and characteristically have a
readable range of 20–100 metres. Frequently, these RFID tags are used to track
larger, more valued assets, such as containers, vehicles or aircraft due to their
effectiveness for tracking objects with a greater distance.

4.1.3 Semi-active/Semi-passive RFID tags

Semi-passive RFID tags have a battery, which can be used to increase transmission
range; for this reason, they are also called battery-assisted tags. These RFID tags
transmit by modulation of backscatter (as do passive tags), but they are still too high
in cost for use with lower value items (Brown, 2007). Semi-active RFID tags, or semi-
passive/battery-assisted RFID tags as they are also called, are now available and
targeted for use in specific applications (Hunt et al., 2007). These types of RFID tags
contain batteries used to support the RFID tag’s sensors and embedded memories
only. When communicating with a RFID reader, the same process used by passive
RFID tags takes place, meaning that energy is generated from the reader antenna in
addition to reflecting a signal back to the reader. As with passive RFID tags, this
process always begins via the reader’s enquiry, followed by the tag’s response. With
passive RFID tags, however, antennas must perform two separate tasks: gathering
energy for the operation of the tag, and transmission of data. Thus, the antenna design
has to be able to balance the performance between these different functions.
Therefore, with battery-assisted RFID tags, which have their own battery for tag
operation, the antenna is enabled to focus solely on data transmission. The result is
that these RFID tags are readable at greater distances, even as far as 30 metres away,
and perform better around liquids and metals compared to regular passive tags. In
addition, time is not required for semi-active RFID tags to acquire energy for activating
the tag chip, meaning that reading speed is also faster, making them more effective at
tracking objects moving rapidly. Table 4 lists five common categories of RFID tags.

39
Table 4: Five common categories of RFID tags (Brown, 2007).

Class Class Layer Name Functionality

1 Identity Tags Purely passive, identification tags
2 Higher Functionality Purely passive, identification + some additional
Tags functionality (e.g. read/write memory)
3 Semi-Passive Tags Addition of on-board battery power
4 Active ‘ad hoc’ Tags Communication with other active tags
5 Reader Tags Able to provide power for and communicate with
other tags i.e. can act as a reader, transmitting
and receiving radio waves

4.1.4 Selection of RFID tags

In terms of selection of RFID tags, there are a number of considerations, which include:

● Frequency Range.
● Memory Size.
● Range Performance.
● Form Factor.
● Environmental Conditions.
● Standards Compliance.
● Read Compatibility.

The RFID tags have been designed to transfer data only if a reader is in range of the
RFID tags, in order to conserve energy and battery life (Brown, 2007). Table 5 below
shows the typical characteristics of RFID tags, examples include:

 Key-ring fob tags.

 Disk tags.
 Wrist band mounted tags.
 Self-adhesive label tags, 30mmx26mm, flat, mounted on plastic.
 Tamper-proof label tags.
 “Credit card” style tags.
 “Laundry” tags (temperature, chemical and heat resistant).
 Tags with a backing label for over-printing.

40
Table 5: Typical characteristics of RFID tags (Author’s completion).

Category Passive tags Semi passive tags Active tags

Power supply External (from Internal battery Internal battery
readers)
Read range Up to 20 feet Up to 100 feet Up to 750 feet
Type of Mostly read-only Read-write Read-write
memory
Cost $.20 to several $2 to $10 $20 or more
dollars
Life of tag Up to 20 years 2 to 7 years 5 to 10 years

4.2 RFID antennas

Generally speaking, RFID readers produce electricity modulated to contain data and
instructions, and feed it to an external antenna. The antenna converts the electricity to
radio waves and broadcasts them. The RFID tag antenna receives the radio waves
and converts them back to electricity used to power the embedded IC chip and to
decode the data and instructions. The RFID tag feeds the electricity to its antenna to
power the circuit and broadcast their coded response, on which the reader reads.
Radio waves travel in a particular direction through space; it can be either horizontally
or vertically polarised as illustrated in Figure 14.

4.2.1 Antenna characteristics

The antenna fundamental characteristics are:

 Impedance.
 Polarization.
 Gain and effective radiated power.
 Bandwidth.
 Appearance.

41
 Circular polarised

Linear polarised

Figure 14: A typical wave pattern of linear-polarized and circular-polarised

antennas (Brown, 2007).

The types of antenna required depend on environment and exact specifications (Jian,
2010). In most cases, antennas with linear polarity are the most common types of
antenna in use, although there are also those with circular polarity (Deavours, 2009).
An antenna that radiates electrical wavelengths with linear polarisation has a much
longer range and higher power levels, which can then produce stronger signals more
capable of penetrating various substances and materials (Deavours, 2009; Jian,
2010). However, these types of antennas are more sensitive to the orientation and
direction of a tag’s position and angle, making some RFID tags harder to read (Nikitin
et al., 2008). In contrast, antennas radiating with circular polarisation do not face this
obstacle in reading tags, although their range and signal strength is much weaker,
relative to power usage. Figure 15 shows a typical antenna that is used in this
research.

Figure 15: A typical antenna. (Impinj company).

4.3 RFID readers

The RFID reader is central to an RFID system, which has the capability of writing and
reading data to RFID tags. It is also becoming more common for RFID readers to

42
exchange information securely by using authentication to validate users via
encryption.

Readers usually have three forms:

1. Stationary readers: These types of readers are usually mounted at gates of

warehouses as shown in Figure 16.

Figure 16: A typical stationary reader (Impinj Company).

2. Handheld readers: These types of readers are convenient, as they can be

used in a more flexible manner as shown in Figure 17.

Figure 17: A typical handheld reader (Bibllotheco Company).

3. Mounted readers: Mounted readers are in between the types mentioned

above. They are usually mounted on a mobile device/system (Henrici, 2008).

Most passive RFID readers, which usually operate under a current of 100–300 mA,
often have the external power due to a much greater quantity of energy consumption.
With this method, an RFID reader is also called an interrogator. An RFID reader is a
device that establishes a wireless connection with RFID tags to capture RFID data.
RFID tags can be accessed by the RIFD reader for a range of tasks in such as filtering,
writing/encoding RFID tags and simple or continuous inventory operations. An
antenna is attached to the RFID reader in order to collect this data and perform these

43
tasks and operations. After a connection is established, the data is streamed to a
computer for processing. RFID readers can be set up in fixed locations, such as within
a warehouse or store, and others may be integrated with portable devices, such as
handheld scanners. In addition, other models can be embedded within electronic
devices and even in vehicles (Brown, 2007). In a warehouse management, the system
needs to identify which item arrives and when to ensure that all items are stored
effectively in the warehouse. The database makes a record and stores the data for
those items that have RFID tags attached to them, allowing those items to be identified
and information data regarding the availability of items stored in the warehouse are to
be known and updated.

4.3.1 RFID reader performance

There are a number of considerations when selecting an RFID reader; importance is

its ability to fulfil the requirements of the system. The performance parameters include
identification range, identification rate, read range, read rate, write range and write rate
as explained in table 6 below.

Table 6: RFID reader performance parameters (Brown, 2007).

Performance Description
parameters
Identification Range The distance at which tags can be identified
Identification Rate The number of tags that can be identified within the
coverage range
Read Range The distance at which tags can be read
Read Rate The maximum number of tags that can be read per
second
Write Range The distance at which tags can be loaded with
information.
Write Rate The maximum number of tags that can be loaded with
information per second

The environment can have major impact on the performance of the RFID reader. For
instance, items like metals and other non-penetrable items can affect the performance
of an RFID reader. Other components of the RFID reader, such as I/O ports, can also
affect the performance of the RFID reader. The types of RFID interfaces include RS-
422, RS-232, RS-485, and Universal Serial Bus (USB) interfaces, which are all
standard interface components for the majority of RFID readers (Jian, 2010; Xiaoyong
et al., 2007). RFID readers become increasingly portable and mobile, enabling more

44
convenience and flexibility when deploying RFID systems, and installations.
Commercial devices might be designed with additional features and components, for
example, input and output ports for annunciator, additional memory and sensors,
which can help to maximise the effectiveness of the RFID reader (Brown, 2007).

4.4 Communication between RFID tags and RFID readers

One of the most important aspects of RFID systems, compared to barcode systems,
is that the communication with the reader does not need a line of sight, or wiring.
Rather, the communication between RFID tags and RFID readers is performed by
electromagnetic means (Henrici, 2008). The radio frequencies used in RFID can be
categorised into three types according to their usage: low frequency, high frequency,
and ultra-high frequency. Frequency refers to the size of the radio waves used to
communicate between RFID systems components. RFID systems operate in low
frequency (LF), high frequency (HF), or ultra-high frequency (UHF) bands depending
on trade-off cost effective solutions. In selecting the right operational frequency for an
RFID system, the specifications of the system need to be assessed carefully as radio
waves have different patterns of behaviour according to the various frequency bands.
One example of this is the ability of radio waves at low frequencies to pass through
most liquids including water, as well as most other materials and substances; these
also operate well in the presence of various metals. At higher frequencies, the
behaviour of radio waves is more like that of light, which tends to be more easily
absorbed by various materials or reflect off other surfaces (Brown, 2007).

Retail inventory management is one of many applications for which UHF is used. In
addition to pharmaceutical anti-counterfeiting systems, LF RFID tags are usually
readable from a distance of 0.3 metres, whereas HF RFID tags are readable up to 1
metre. In contrast, high data transfer rates are achievable using UHF waves, although
the rates of performance in environments containing liquids and metals are
significantly reduced, and data transfer becomes practically ineffective (Hunt et al.,
2007). Nonetheless, there was an ongoing development of passive RFID tags that
work in UHF, which continues to gain interest (Ting et al., 2012). In terms of the
motives for using UHF-RFID tags in supply chain management, as opposed to LF
RFID tags or HF RFID tags, it simply comes down to cost effectiveness and the read-
range of the RFID tags, with commercial vendors offering relatively simple and

45
inexpensive RFID tags in the UHF band. UHF RFID is the most sensitive to radio wave
interference compared to LF and HF RFID tags; however, many UHF products, such
as RFID tags, antennas, and readers, are designed to perform even in more
challenging environments (Impinj, 2015).

4.5 Electromagnetic field

There are two kinds of fields used for data transmission and powering the tags:

 RFID Near Field Communication (NFC).

 RFID Far Field Communication (FFC).
There are certain key criteria in which NFC and FFC differ. The range for RFID tag-
reading with NFC reading RFID tags is no more than a metre away, while FFC is able
to read RFID tags from distances of 10–12 metres away. Other differences include the
way the data is stored and transferred. Identifying the differences in terms of
functionality will help in selecting the correct capabilities and tolerances for meeting
the required specifications (Nikitin & Rao, 2008). The type of antenna required for
operations is also dependent on the distance involved in reading the RFID tags, which
is called the read-range. There are “near-field” or short range antennas, for which the
read-range is less than 30–35 cm, and “far-field” or long range types, which can have
a read-range of several metres, or perhaps even twice or thrice that distance in
optimum conditions. The maximal field strength allowed depends on national
regulations. It is limited for electromagnetic compatibility (EMC), i.e., for avoiding
disturbing other systems and to prevent harming the environment (Brown, 2007).

4.5.1 Near Field communication vs Far Field communication

RFID functions at different frequencies, including low, high and ultra-high frequencies.
The implementation of the frequency is to determine the distance in which the RFID
tags can be read, how many RFID tags can be read at a time, how fast these tags are
read and how the performance will impact on its application environment (Chawla,
2007). Table 7 shows the individual performance between near-field communication
and far-field communication in terms of capabilities, strength, weakness, and
application suitability. Figure 18 shows different frequencies between the near-field
communication and the far-field communication.

Table 7: Comparison in performance between NFC and FFC (Author’s completion).

46
 Near Field Communication Far Field Communication

Read range Up to 1 meter Up to 10- 12meters

Security Data is stored in a tag itself Data ID is stored and server is
required
Range control Fixed Controllable
Transfer data Low Fast
Waiting time More Less
Design Standardised Tamper proof
Queue One by one reading from close Automated reading from controlled
management distance distance
Reading No multiple reading capabilities Multiple reading capabilities
Cost Cheaper Moderate

RFID System

Near Filed Far Filed

125 KHz 315 MHz

13.56 MHz 433 MHz
860-870 MHz
902-928 MHz
2.40-2.48 GHz

Passive Passive Active

Read range > 50cm Read range > 3-4m Read range > 100m

Figure 18: different frequencies between NFC and the FFC. (Author’s completion).

4.6 RFID controllers

Controllers are devices that allow a group of RFID readers to communicate with a
computer system, i.e., the data server. With the proliferation and increased complexity
of RFID systems, it has become necessary to introduce a device that handles the
communication between many RFID readers and a host computer, where the
information gathered from the RFID system is stored. Reva Systems, a leader in the
integration of RFID within current network infrastructure, designed an architecture that
simplifies the network management of RFID readers with a product called Tag
Acquisition Processor (TAP) (Brown, 2007). The aim of TAP is to move away from the

47
reliance on middleware by having a controller that provides an efficient integration with
enterprising applications.

4.7 RFID software

There are two categories of RFID software as follows:

1. RFID middleware.
2. Application software.

4.7.1 RFID middleware

RFID middleware handles data and communicates with RFID reader devices.
Middleware, as the name suggests, provides a middle level event filter. For almost all
RFID management events, repetitive read cycles are needed because of reading
inaccuracies. High-volume raw data are processed by the middleware in detailed read
cycles with results that contain disinteresting streams of data, which are useless to its
end-users. Processing the data received from RFID reader devices provides the user
applications with more clear and comprehensive results at the event level (Brown,
2007; Henrici, 2008). In addition, standard application level API is provided, with event
level information being sent over a standardised interface via a service connecting it
to higher level applications. The gathered information can also be stored by
middleware, such as with a backend database that stores RFID tag information. The
RFID middleware is logically positioned in the midway between the real business-end
applications and the RFID hardware infrastructure. The middleware usually operates
the RFID reader devices via the server that connects it to the RFID readers or to the
RFID reader network. This allows direct communication and control between all
devices, enabling event information to be prepared and processed, in a standardised
format at the higher level, for its business-end applications. This standardisation is
defined in EPC Application Level Events specification (ALE), which describes and sets
out logical architecture and the implementation of RFID middleware. Numerous
commercial RFID middleware products compatible with EPC ALE are available in the
market (EPC Global, 2005). The RFID middleware is located between RFID hardware
infrastructure and real business applications. Middleware normally works on the server
that is connected to the RFID reader. The middleware communicates and manages
the event data/information that is being created by an RFID device in a standardised

48
format in order to match business applications at a higher level. It is suggested that
middleware must work within the constraints of corporate standards and security
policies, and interface with warehouse management system (WMS), however, each
company has its own unique set of conditions and requirements. The main tasks of
the RFID middleware include:

● Device management.
● Data collection and integration.
● Data structuring.
● Data filtering and routing.
● Line coordination and control.
● Visibility and reporting (both business information and device
performance).
● Track and trace applications for recall and shipping (Park & Yu, 2008).

4.7.2 Application software and implementation for RFID systems

RFID systems require application software for RFID implementation. Brown (2007)
stated that some companies used standard ERP applications from vendors such as
SAP, JD Edwards, Oracle, PeepleSoft, or Microsoft to manage their supply chain
activities. Also, other companies utilise manufacturing execution systems (MES),
warehouse management system (WMS), supply chain management systems (SCMs),
and finite forward scheduling (FFS) packages. The software needed for an RFID
system to function effectively involves RFID reader control and application software,
which is frequently referred to as middleware, which enable the RFID reader operation
and data communication that is necessary for operating the RFID system if the
application is a new system, then the RFID system can be embedded/integrated
initially. On the other hand, if the RFID system is integrated with the existing system,
it is very important to define and decide how to interface with it for integration. An RFID
reader can read multiple RFID tags at the same time, this is called batch reading. The
RFID system also needs an effective mechanism to deal with the enormous amounts
of data being processed per second. All records of RFID tags are stored in the data
server where the data volume, data capturing speed, and complexity of the data being
captured by the RFID readers can make the process of the data for storage very
challenging. The WMS also maintains instant inventory updates via the control unit

49
that continuously transmit data over a LAN using middleware, which is the layer of
software or programmed applications that translate commands and data between an
RFID reader and the organisation’s enterprise system. Databases are compiled in
terms of identification and availability of each product and other data variables.
Advantages of RFID include the ability to obtain more accurate product information
and data transfer, enhance data security and control, and ensure adequate speeds of
data-processing. Pallets and other equipment passing through doors and access ways
can also be monitored using RFID providing sufficient coverage of the plant.

4.8 RFID cost elements

There are three cost categories (i) Hardware cost, (ii) Middleware cost, and (iii) Service
cost. Hardware cost category covers the costs of tangible elements of the RFID
system, such as RFID tags and RFID readers. Hardware cost is considered usually in
the preliminary planning stages and it is important to note that economic models
should also consider other cost categories. Justification of hardware costs is not
sufficient for implementation decision, although hardware vendors may suggest
otherwise. Service cost, such as business process redesign cost, and configuration
cost, require an elaborate study on the firm and supply chain specific requirements. A
whole range of cost elements may emerge depending on the context (Baysan et al.,
2013).

50
Summary

In this section, key components of the RFID-enabled inventory management system

are described in terms of functionality and implementation. This includes RFID
database, RFID tags and an RFID reader, RFID antenna, bandwidth/connectivity of
RFID system, connectors/cables of the RFID systems, RFID controllers and RFID
software. To synchronize products and information flow, both RFID-based inventory
management system and the warehousing control system need to be integrated in
order to communicate effectively through a developed interface. The methodology for
systems integration including hardware and software were identified and proposed as
part of this research study.

51
Chapter 5 The proposed RFID-based inventory management system and
measurement

5.1 Mechanism of the proposed RFID-based automated warehouse system

Figure 19 illustrates the key components of the RFID-based management system for
the proposed automated warehousing system. At the gate of the warehouse
entrance/exit, it is equipped with an RFID reader (shown at A) which collects RFID
information data of RFID-tagged goods for each incoming and outgoing lorry that
passes through the gate. After the process of unloading and unpacking goods if
applicable, each item will be contained in a tote entering into a storage conveyor,
namely a storage rack (shown at B1). When a specific item is demanded, it can be
transferred automatically from a storage rack onto an output conveyor (shown at B2)
by a pusher device (shown at B). Each pusher also contains an RFID reader with its
own antenna that receives wireless signals sent from RFID tags attached with RFID-
tagged items in storage racks. Collection of inventory data by the RFID reader is
updated instantly by a warehouse management system through a controller which
transmits the collected data via a middleware (shown at D). The middleware is the
software translation layer between an RFID reader and the warehouse management
system. Database of the warehouse management system contains records, which
include data in identification, availability and other use-defined information of each
item stored in the warehouse. Once an in-store item is ordered, the RFID-based
inventory management system has capability to carry out an automatic check on
information data of the item in the warehouse database. Once the ordered item is
identified by the RFID-based inventory management system, a pusher is activated by
a PLC (programmable logic controller) to push the selected item in a tote onto an
output conveyor. The item will then be transported by the output conveyor and it travels
along an RFID-guided route to a specified destination (i.e., a collection point) for
packaging (shown at C). The RFID-based warehouse inventory database will then be
updated as soon as this ordered item is shifted out of the distribution centre in a lorry
passing through the gate of the warehouse entrance/exit. The whole process is
performed automatically without any human intervention apart from unpacking,
labelling and packing operations in the warehouse.

52
 RFID
Pusher
Reader
Open area for packaging/
labelling/ unpacking. C
Tagged
Web Tagged
items server
B items

Middleware
Database B2
storage
Conveyor
D User application
RFID Controller
Tagged
RFID items
reader gate B1
Exit A

A. Portal reader. B. Storage area. C. Open area for packaging/ labelling/

unpacking. D. Data Centre.
Figure 19: Key components of an RFID-based automated warehouse system.

5.2 The RFID-enabled automated storage and retrieval system (AS/RS)

Figure 20 illustrates an automated storage and retrieval rack (AS/RR) as one of the
core components of the proposed RFID-based automated warehousing system. The
module is designed as a standardised element for manufacturing and assembly,
although each module can be of different sizes and arrays in a module that can be
configured easily in many different ways, i.e., capacity of a warehouse are adjustable.
The module is comprised of two types of powered conveyors aligned next to one
another; these are input conveyors (storage racks) and output conveyors. The
operation of both conveyor systems is controlled by a PLC that communicates with
mounted sensors via a local area network (LAN). Within the RFID-inventory
management system, a chosen SKU can be released by the mechanical system of
AS/RR based on a number of assignments polices or rules. These include, for
example, the rule of being nearest to a collection point and/or a pusher which is free
or adjacent to the chosen SKU and so on, which are explained (in chapter 6 sections
6.4.1).

In the warehouse, once an item (or a tote that contains identical items) is attached with
an RFID tag, it can be tracked and manipulated by the RFID-based inventory

53
management system. An RFID tag has a unique identification and other user-defined
information in association with a status of each item. It is composed of two main
components which are an antenna and a computer chip. The computer chip stores
data and the antenna allows data communication between an RFID tag and an RFID
reader through a wireless signal transmission. A typical RFID reader is a
microcontroller-based radio transceiver that powers an RFID tag using the time-
varying electro-magnetic field (EMF) generated from an RFID antenna. Two types of
RFID tags (active and passive RFID tags) can be used depending on a range in RFID
reading performance at a location of an AS/RR. The RFID information data gathered
by an RFID reader can be transferred to a host PC database for data processing and
storage (as illustrated in Figure 19). By using RFID tags, a SKU can be distributed
randomly at any location of an AS/RR wherever a place is available for incoming goods
as illustrated in Figure 21. Thus, a SKU can also be dispatched randomly at varying
locations for outgoing goods in the warehouse. This significantly facilities operations
of storage, retrieval and replenishment, and improves capability, flexibility and
responsiveness of the warehousing system to store and dispatch an item in/from an
AS/RR.

The output conveyor system Items enter onto

a storage rack

Tagged
items

Spiral conveyors

Pusher

Storage rack

Output to collection points

Figure 20: Structure of the automated storage and retrieval system.

54
 Random location and distribution of storage and retrieval of
items throughout the RFID-enabled warehousing systems

Output conveyor
A pusher Storage module array

Transfer conveyor system

A collection point

Figure 21: Random storage and retrieval locations of SKUs.

5.3 Measurement of the storage and retrieval mechanism

Figure 22 illustrates the key components and corresponding geometric parameters of

the proposed storage and retrieval system (as illustrated in Figure 20): 1) output
conveyor; 2) storage rack containing items; 3) height of a storage rack H; 4) length of
a storage rack L; 5) depth of a storage rack D; 6) pusher; 7) length from an end of an
output conveyor to an entrance of a spiral conveyor; 8) entrance to a spiral conveyor;
9) length of a single spiral conveyor; 10) a spiral conveyor; 11) length from an end of
a spiral conveyor to a collection point. The aim of the following work is to determine a
total travel time an item needs from the moment when a pusher device is activated to
push a selected item in a tote onto an output conveyor to the moment this item travels
down to a collection point. Figure 23 shows a two dimensional cross-section diagram

55
 based on Figure 22. The pusher can move simultaneously in both horizontal and
vertical directions from one location (i, j) at where the pusher currently stays to another
location (my, nx) at where an identified item is selected to be pushed onto the output
conveyor.

(7)

(8)
(9)

(2)
(1)

(10)
(3)
(6)

4
(11)
Figure 22: Geometric parameters corresponding to the key components of the
storage and retrieval system.

Y
Row Column
Column
𝐼
Storage
(𝑚𝑥 ,𝑛𝑦 )

Item location
𝐿𝑠𝑏 H
𝐿𝑑

P (𝑖, 𝑗)
X

L Row
Figure 23: Geometric parameters in a storage rack.

56
5.4 Travel time

The following assumptions and notation are used when two cases of the velocity
profiles for the pusher:

- a: acceleration/deceleration rate of the travelling S/R pusher.

- v(t): velocity of the travelling S/R pusher at time t.
- Vmax: maximum velocity of the travelling S/R pusher.
- d(t): moving distance at t.
- ta: time necessary to reach the peak velocity.
- T: total travel time of the pusher.

There are two cases of velocity profiles that are used to compare the total travel time
of the pusher. Case I where pusher has three different motion phases – acceleration,
constant speed and deceleration. In case I, pusher first starts to accelerate to obtain
the maximum speed Vmax and then it travels with that maximum speed for certain time
and then decelerate to reach the final destination as illustrated in Figure 24. Case II
where pusher travels with a constant speed of Vmax throughout whole journey. as
illustrated in Figure 25, where the traveling S/R pusher V(t) is:

at t   0, ta  
 
V (t )  v max t   t a , T  t a  (1)
 
a(t  T ) t  T  ta , T  
Thus,

T 2
vmax
d (t )   v(t )d (t )  vmax  (2)
0
a

Equation 1 and 2 describe the velocity-time and distant-time relationship of the pusher
when case I motion is applied on the pusher.

57
Velocity

𝑉 𝑚𝑎𝑥

Time
0 𝑡𝑎 𝑇 − 𝑡𝑎 𝑇

Figure 24: Velocity-time relationship of the travelling S/R pusher for the case I motion
where acceleration, constant speed and deceleration is utilised.

- S/R pusher traveling for case II (Figure 25).

Velocity

𝑉 𝑚𝑎𝑥

Time
0 𝑡𝑎 𝑇 − 𝑡𝑎 𝑇

Figure 25: Velocity-time relationship of the S/R pusher travelling- case II.
The following assumptions are used in this case:

1. A pusher is located at the centre (ic, jc) of a storage rack; this is a default location
of the pusher to be activated.
2. The pusher is capable of moving simultaneously in both vertical and horizontal
directions at an estimated speed, this implies that the pusher can move along
the linear route Ld to a specified location before pushing a selected item from

58
 a storage rack onto and output conveyor.
3. Each item in a tote is pushed onto an output conveyor instantly without any
delay.
4. The specification of the pusher (max. velocities in horizontal and vertical
directions; and acceleration and deceleration rates) as well as the length (L)
and height (H) of the SR are known.
5. The L and H of the SR are long enough for the S/R pusher to reach max.
velocity from the default location.
6. A SKU can also be dispatched randomly at varying locations for outgoing goods
in the AS/RR.
The following notations are used in this model.

- I: an item located at a row mx and a column ny i.e., (mx, ny) in an AS/RR.

- P: a pusher at a location (i, j).
- Lsb: length of a tote in a storage room of an AS/RR.
- Lsc: length from a position of a selected item in a tote to be pushed onto an output
conveyor to a position where this item in a tote travels at an end of an output
conveyor.
- Lss: length of a single spiral conveyor.
- Lsp: length from an end of a spiral conveyor to a collection point.
- Ld: distance between a pusher and a selected item.
- Vc: speed of an output conveyor.
- Vp: speed of a moving pusher along Ld.
- Vpp: speed of a moving pusher to push an item onto an output conveyer.
- Vs: speed of a spiral conveyor.
- ci: chosen item, (where ci = 1, 2….., n).
- Tm_ci: time needed for a pusher to move to a selected (chosen) item ci.
- Tp_ci: time needed for a pusher to push a selected item ci onto an output conveyor.
from an input conveyor (i.e., a storage rack).
- Ts_ci: travel time of a selected item ci in a tote along an output conveyor to a spiral
conveyor.
- Tsc_ci: travel time for a selected item ci in a tote to move from the top level to the
bottom level of a spiral conveyor.
- Tse_ci: travel time of a selected item ci from an end of a spiral convey to a collection
point.

59
- n:number of chosen items.
- Ttt_ci: total travel time of each of selected items from the moment when the pusher
is activated to push the selected item to the moment when the selected item arrives
at a collection point.
- Ttt: least travel time of one of selected items from the moment when the pusher is
activated to push the selected item to the moment when the selected item arrives
at a collection point.

Illustrated in Figure 22, assuming an item is located at (mx, ny) and a pusher has a
random location at (i, j) in a storage rack. Thus, a distance Ld between the activated
pusher and the selected item is given by:

Ld  (mx  i)2  (ny  j )2 (3)

Therefore, the travel time Tm is given by:

Ld
Tm  (4)
VP
Where, Vp is a speed of the moving-pusher along Ld. We define Tp is a travel time
from the moment when the pusher starts to push a selected item to the moment that
the selected item has been pushed onto an output conveyor, it is given by:
D
Tp  (5)
VPP

Where, D is a depth of the storage rack at which the selected item is located. Vpp is a
constant speed of the moving-pusher to push the selected item onto the output
conveyer. Knowing Lsc refers to a distance from the centre of the tote at the end of the
conveyor. Lsb is a length of a storage room containing each item in a tote, ny refers to
the number of columns in row mx , Vc is the constant speed of the output conveyor.
Thus, 𝑇𝑠 is given below:
Lsb  (ny  0.5)  Lsc
Ts  (6)
VC

The item travels down through a powered spiral output conveyor from the top level to
the bottom level towards a collection point where the item is collected for packing.
Thus, Tsc is calculated by:

60
 mx  Lss
Tsc  (7)
VS

Where, Lss is the length of the single level spiral conveyor. mx refers to the number of
rows in column ny. As Tse is a travel time that the item needs from the end of the spiral
conveyor to a collection point. LSP is a travel distance between the end of a spiral
conveyor system and a collection point. Therefore, the travel time between the end of
a spiral conveyor and a collection point is given by:
LSP
Tse  (8)
VC

Hence, the total travel time Ttt_ci for the selected item is obtained by:

Ttt _ ci  Tm _ ci  Ts _ ci  Tsc _ ci  Tp _ ci  Tse _ ci (9)

T Where, ci= (1, 2, ……n). The selected item that needs a minimal travel time Ttt ,
which is identified by the RFID-inventory management system throughout the storage
and retrieval system, can be obtained by
Ttt  min (Ttt1 , Ttt 2 ,................,Ttt _ ci ) (10)

The longest travel time Tmax of the one of selected items to a collection point can be
obtained by
Tmax  max[Ttt _ ci , Ttt _ ci 1 ,..........., Ttt _ ci n ]
(11)

After an order for multiple items are made by a customer, the RFID-inventory
management system needs to determine which item has a longest travel time and
subsequently issue a priority to this item to be dispatched first. Figure 26 illustrate the
multi-directional trajectory of a pusher which travels to a selected item. Assuming that
an item has a random location (mx, ny) and a pusher is located at the centre (ic, jc). of a
storage rack; this is a default location of the pusher to be activated. The pusher P can
move simultaneously in both horizontal and vertical directions from the centre location
(ic, jc) at where the pusher currently stays to a location (mx, ny) at where the selected
item awaiting to be pushed onto the output conveyor by the pusher mounted in AS/RR.
As stated previously, when an order for multiple items is executed, the pusher can
only push one item at a time according to the estimated travel time determined by
RFID-based inventory management system (illustrated in Figure 44, chapter 7).

61
Therefore, it is desired to determine an overall estimated waiting time for an order that
contains multiple items from the moment when a pusher is activated to push the first
selected item with a longest travel time to a specified collection point to the moment
that the last selected item with the least travel time arrives at the same collection point.
With this approach, it gives a minimum waiting time for all the items to meet at the
specified point.
The following notations are used.

- Pc: a pusher at a centre location (ic, jc)

- V1 : a constant speed of an output conveyor.
- V2: a constant speed of a spiral conveyor.
- 1B : identified items in each row at different columns of an AS/RR.
- 1L : identified location of each of ordered items in each row at different columns of
an AS/RR.
- Tmax: longest travel time of the one of the selected items.

The farthest storage location

Y point to the collection point
(ic, jc)
V1

(mx, ny)

𝑃𝑐

V2

(1,1) (1, ny)

X
The closest storage location
A collection point Row
point to the collection point

Figure 26: Multi-directional trajectory of a pusher which travels to selected items.

62
5.5 Indicator function

In mathematics, an indicator function is a function that defined by a set of 𝑥 that

indicates the membership of an element in a subset B of x, assuming that a value 1
for all elements of B and a value 0 for all elements of X not in B.

The indicator function of a subset B of a set X is a function defined as below

1 if x  B
1 ( x)   0
B
if x  B
(12)

Let B be the set of random locations (mx, ny) for each of items in AS/RR

1 if (mx , ny )  B
1B(mx , ny )   0 if (mx , ny )  B
(13)

As an example, if 1B (mx, ny) =1, it indicates that an ordered item is identified by

the RFID-inventory management system, i.e., this ordered item is available at this
particular location (mx, ny). If 1B (mx, ny) =0, it indicates that an item is not ordered.
Figure 27 illustrates this process. Assuming that there are multiple items ordered by a
customer, the inventory management system will then search and identify a location
of each ordered item in an AS/RR. The system will search and identify a location of
each ordered item from location (1,1) to location (1, ny) in a row across each storage
rack of an AS/RR as shown in Figure 26. Figure 27(a) shows the process of searching
and identifying locations that contain ordered items from left to right in row 1 of the
storage rack. Figure 27(b) shows a number of identified locations containing ordered
items in a storage rack by issuing a value 1 for each of the ordered items and a value
0 for a non-ordered item at a location (mx, ny). Figure 27(c) shows the final result of
selected locations which contain ordered items in this storage rack.

63
 𝑌
1𝐵 (𝑚𝑥 , 𝑛𝑦 )
(𝑚𝑥 ,𝑛𝑦 )
1

𝑋
(a) 0
𝑌 1𝐿 (𝑚𝑥 , 𝑛𝑦 )
Three locations of selected items
(𝑚𝑥 ,𝑛𝑦 )
1 1 1
0 0 1 0
𝑋
(b) 0
𝑌 𝑛
(𝑚𝑥 , 𝑛𝑦 )
1
1 1 1
(c) 𝑋
0

Figure 27: Mechanism of searching and selecting an ordered item in a storage rack.
Hence, a total number of ordered items n can be obtained by

m x , n y ) 1L ( m x , n y )
n  1B ( (14)

Figure 28 illustrates the routing algorithm for a pusher to be given the priority to travel
to one of selected items. Let us define Ttt which refers to the total travel time needed
for a selected item to travel from the moment a pusher is activated to push the ordered
item to the moment this item arrives to a specified collection point. Assuming that a
number of items n are ordered from the warehouse, these items are randomly placed
at different location ci (where ci= 1, 2, …..n) in the AS/RR. The total travel time for
each of selected items is denoted by Ttt_ci.The RFID-inventory management system
calculates an estimated travel time Ttt_ci for each of the selected items to a specified
collection point in order to determine the item with a longest travel time to be given a
priority to be pushed onto the output conveyor by the pusher. The system will then
repeat the same process by determining the second item with a longest travel time
among the remaining items until the last item with a least travel time to be pushed onto

64
the output conveyor. With this approach, the pusher can also travel through in an
optimal route to ensure that all the selected items will arrive and meet at the specified
collection point with a minimal waiting time for packers to receive these items.

Number of ordered items (n)

Calculate a travel time of the pusher to

each selected item

A
NO
Ttt_ci > Ttt_ci+1

B
YES
NO
Ttt_ci < Ttt_ci+1 Ttt_ci+1
C
YES
Ttt_ci= Ttt_ci+1
Ttt_ci

Default to select the item

that is closest to the pusher 1 2 3 …..... n

Figure 28: The routing algorithm for a pusher to travel to each of selected items.

65
Summary

This chapter describes the mechanism of the proposed RFID-based automated

warehousing system and its key component of AS/RR; one of the core components of
the proposed RFID-based automated warehousing system is the design of the
automated storage and retrieval system AS/RS. The module was explained and
illustrated in details for scheduling job priorities to selected items to be dispatched from
the warehousing system travelling to a collection point. Thus, the aim of this work is to
determine a total travel time an item needs from the moment when a pusher device is
activated to push a selected item in a tote onto an output conveyor to the moment this
item travels down to a collection point. Indicator function was identified to in order to
search and select an ordered item in a storage rack. With the developed routing
algorithm approach, it gives a minimum waiting time for all the items to meet at the
specified point.

66
Chapter 6 Integration of the RFID-enabled warehouse management system

6.1 Key components used for integration of the RFID-enabled warehouse

management system

Figure 29 shows the key components used for an experiment towards the integration
of the RFID-enabled warehouse management system. A typical RFID reader was used
and it is a microcontroller-based radio transceiver that powers RFID tags using the
time-varying electro-magnetic fields (EMF) generated from an RFID antenna. Two
types of RFID tags (active and passive RFID tags) can be used for this experiment
depending on a range of RFID reading performance at a location of an AS/RR. An
electromagnetic field’s characteristics change depending on the distance from the
antenna. This varying field is broken up into two parts: the near field communication
(NFC) and the far field communication (FFC). An RFID tag is composed of two
essential components: an antenna and a computer chip. The computer chip is used to
store data while the antenna allows data communication between the tag and the RFID
reader through a signal transmission. RFID tags have their unique identification (UID),
which is used to uniquely identify an item each RFID tag is assigned to. The collected
RFID tags can be read by the reader and sent back to the host PC for processing.

Arduino and Microsoft Visual Studio environment software are used to allow users to
perform a group of coordinated functions, tasks, or activities.

MATLAB is a foundation for all other MathWorks products providing a high level of
programming language and interactive technical computing environment and
functions for algorithm development, data analysis and visualization. In this study
MATLAB was used to demonstrate the data process output of ordering items using
the developed algorithm.

67
 Arduino software

Management system

Arduino Uno RFID tag

RFID reader RFID tag Impinj RFID reader Antenna

(a) Near Field Communication (b) Far Field Communication

Figure 29: Key components used in this experiment.

6.2 RFID near-field communication

Near field communication (NFC) is a set of standards for smartphone and similar
devices to establish radio communication with each other. Applications include
contactless transactions, data exchange, and simplified setup of more complex
communications such as Wi-Fi. Figure 30 shows Arduino which is an open-source
platform and it can be used to develop stand-alone interactive objects or can be
connected to software on a computer (e.g. Flash, Processing, MaxMSP). The Arduino
has the ability to be installed to any type of operating system and it can be powered
and programmed via the built in USB connection. Arduino UNO contains; 6 pins used
for pulse width modulation (PWM) as outputs and 6 pins used as analogue input.
Arduino UNO also includes a 16 MHz crystal oscillator, a USB connection, a power
jack, an ICSP header, and a reset button. Table 8 shows specifications of Arduino
UNO.

68
 General I/O
Reset button

USB connector ATMega 328 MCU

7-12VDC input Analogue-to-digital converter input

Power pins

Figure 30: Arduino UNO (Arduino Company).

Table 8: Specifications of Arduino UNO.

Microcontroller ATmega328
Operating Voltage 5V
Input Voltage 7-12V
(recommended)
Input Voltage (limits) 6-20V
Digital I/O Pins 14 (of which 6 provide PWM output)
Analogue Input Pins 6
DC Current per I/O Pin 40 mA
DC Current for 3.3V 50 mA
Pin
Flash Memory 32 KB (ATmega328) of which 0.5
KB used by boot loader
Clock Speed 16 MHz

6.2.1 Recognition between an RFID tag and an RFID reader

Figure 31 shows a flowchart of reading activities between an RFID tag and an RFID
reader using the ID-12 reader model. Initially, the device will determine the input and
output pins to be used in order to perform the task of reading an RFID tag, and then
set up the serial connection to the RFID reader module. Once the Wi-Fi connection is
established, the RFID reader waits to obtain the unit code, as it starts to read tags
each UID tag will be transmitted to the server.

69
 Start

Determine I/O pins

NO
Is Wi-Fi
connected?

YES
Connect and read tags

Transmit data to the database

Finish

Figure 31: Reading activities between an RFID tag and an RFID reader.

The ID-12 reader module was used because it meets the requirement that includes
the actual size of the RFID reader and internal antenna with a range of 12cm. It
requires between 2.8-5 volt of power supply, a ground pin and a digital pin (on the
Arduino) that will be utilised for serial connection. The data retrieved from RFID tag ID
is programmed to transmit instantly to the server for fast response by RFID ID-12
reader, as storing data to the tag memory will lead to delay/bugs. The RFID (ID-12)
reader is a low frequency (LF) RFID with 125 kHz frequency, it supports ASCII,
Wiegand26 and Magnetic ABA track2 data formats. In this experiment, an RFID reader
with ISO/IEC 18000 Part 3 Standards as shown in Figure 32.

70
 Figure 32: The RFID Reader (ID-12) (Author’s completion).

6.3 RFID far-field communication

Figure 33 shows key components for a pilot test of the RFID inventory management
system, which has a capability of interacting with the mechanical control system of the
proposed RFID-embedded automated storage and retrieval rack (AS/RR). The system
hardware includes a host computer, an ultra-high frequency (UHF) Impinj RFID reader,
an Impinj far-field antenna, and a number of passive tags. The RFID reader is attached
with the antenna in a range of 902~928 MHz (using the EPC global Gen2 ISO18000-
6C standard) in a direction of 70° to capture data transmitted from the passive RFID
tags and these gathered data are stored in a laptop for data processing (Liu et al.,
2014). The RFID-based inventory management system was developed under the
Imping Octane SDK, which is a package for configuring settings of the RFID reader
using a Low Level Reader Protocol (LLRP). LLRP offers a high-level control in reader
settings, tag query and tag-writing operations. The RFID reader API (application
program interface) package provides a common programming interface to allow an
interaction between the RFID hardware and the RFID application software hosted in
an MVS (Microsoft Visual Studio) environment. The software also defines a high-level
object-oriented interface that permits the communication with the RFID reader.

71
Figure 33: Key components of the RFID inventory management system for the pilot
test.

The MVS supports a NuGet which is an open source package manager providing a
number of application tools including the entity framework, which is an object relational
mapping (ORM) mechanism. Within the ORM mechanism, it can create classes based
on database tables in which each entity framework allows a web application to
communicate with the database. The inventory management system contains a
transaction database using the Microsoft SQL server. The database can automatically
update an existing location of an RFID-tracked item at any time when a change of it
takes place. As an example, when an RFID event (e.g. a reader reads an RFID tag)
occurs, the database automatically updates the record of the collected information
relating to this event. The inventory management system then runs the associated
program using the selection algorithm to determine the updated location of an RFID-
tracked item in the warehousing system and the location of this item can be displayed
in the GUI (graphical user interface) (as shown in Figure 44 in chapter 7). The GUI
was developed in MATLAB and it allows for visualizing changing locations of any
possibly selected item in the warehousing system. MATLAB toolbox supports
communications using the open database connectivity (ODBC) that can link with any

72
compatible database such as Microsoft SQL. An ODBC is a middleware application
program interface that allows an access of the management system database.

6.4 Database of the RFID inventory management system

The database of the inventory management system is the core of RFID systems. The
database can be implemented using Microsoft SQL, Microsoft SQL, Sybase, Oracle,
DB2, MySQL or MS-Access. In this study, the database of the RFID-inventory
management system was created using Microsoft SQL. The database records contain
data of each item attached with an RFID tag indicating the identification and availability
of this item stored in the warehouse. Table 9 shows a list of records of searched items
which contain RFID-inventory information within the SQL database.

Table 9: Part of information data within the SQL server.

Type of Item Expiry RFID Tag id Label

item location date number
'Milk' (40, 50) '20160201' '3008-33B2-DDD9-0140-0000- 1111119
0001'
'Milk' (50,30) '20160222' '3008-33B2-DDD9-0140-0000- 1111118
0000'
'CD' (90,60) '00000000' '3008-33B2-DDD9-0140-0000- 1111120
0002'
'Mobile' (80,20) '00000000' '3008-33B2-DDD9-0140-0000- 1111121
0003'
'Bread' (70,10) '20160203' '3008-33B2-DDD9-0140-0000- 1111122
0004'

6.4.1 The data processing algorithm using MATLAB

As stated previously, the RFID-inventory management system has capability of

performing an inventory searching process of ordered items by executing a set of pre-
defined selection rules, which include an availibilty check of each of ordered items,
locations of ordered items, expiry dates of each of ordered items if applicaple, a
shortest path for the ordered item traveling to a specified collection point and other
user-predefined selection criteria. Figure 34 illustrates the inventory selection process
of ordered items to be dispatched by the warehouse. If an item is selected from a
group of the same type of items stored in multiple locations in the warehouse, the
RFID-inventory management system will issue a priority based on the pre-defined

73
selection rules to be given to a selected item and initialise a demand to push the
selected item onto the output convyeor (as illustrated in Figure 20 in chapter 5). To
schedule a job priority for the selected item to be dispatched from the warehousing
system, an algorithm was developed to seek an optimal solution for selecting an item
which has a priority over other items of the same type. Table 10 shows part of the
programming codes which demonstrate the data process of ordered items using the
developed item selection algorithm (Other codes can be found in Appendix C).

User

Receive information of ordered

items

YES
Check if each ordered
item has an expiry
date?
NO
NO
Are there multiple locations
containing each item?
Check the varying
YES expiry dates? YES
NO
Identify the item Push the item
with a priority of a with a priority of a
least travel time to nearest expiry
a collection point date

To activate the
pusher to push the
selected item onto
the output conveyor

Display RFID codes for the selected items

Figure 34: The RFID-data processing algorithm.

74
Table 10: Part of MATLAB programming codes for data processing.
================================================================
Data Processing
disp('Please Input The Name of Ordered Item!')
OrderedItem='Milk '; % Change the name of ordered item.
Find and save the location information which contains the ordered item
No=0; % Starting number of ordered item
Location=[]; % For storing the location information of ordered item
Calculate the computation time
Tm=sqrt((PP(1)*Hf-Location1(1)*Hf)^2+(PP(2)*L-Location1(2)*L)^2)/Vp;
Tp=D/Vpp;
Ts=((Location1(2)-0.5)*L+Lsc)/Vc;
Ti=(rack-Location1(1))*Lss/Vs;
Te=Lsp/Vc;
Tt=Tm+Tp+Ts+Ti+Te;

6.5 Standardisation of RFID Technology

RFID standards are guidelines or specifications for all RFID products. Standards
provide guidelines on how RFID systems work, what frequencies to operate at, how
data is are transferred, and methods in communication between the RFID reader and
the RFID tag. The UHF frequency band covers the range from 300 MHz to 3 GHz.
Systems comply with the UHF Gen2 standard for RFID use the 860 to 960 MHz band.
While there is some variance in frequency from region to region, UHF Gen2 RFID
systems in most countries operate between 900 and 915 MHz. Passive UHF RFID is
currently the only type of RFID to be regulated by a single global standard. This
standard is called EPCglobal UHF Gen 2 V1, or just UHF Gen 2. UHF Gen 2 defines
the communications protocol for a passive backscatter, reader-talks-first radio
frequency identification (RFID) system operating in the 860 MHz - 960 MHz frequency
range. EPCglobal certification testing includes conformance testing, which ensures
that RFID products are compliant with the UHF Gen2 standard, and interoperability
testing, which makes sure that all aspects of the tag reader interface are properly
designed to interoperate seamlessly with other Gen 2 certified products. Table 11
shows the product code for RFID components that used in the research work.

Table 11: RFID components product code (Arduino Uno company, Impinj company).

Product type Product Code

Fixed Readers-Impinj Speedway Revolution IPJ-REV-R220-EU11M1

75
 MTI Far Field Antenna THI004
Label RFID Tags-Casey CON008
EPC Global Class 1 Gen 2 (ISO 18000-6C)
Compliant

Arduino UNO A000066 (TH); A000073 (SMD)

The ID-12 reader module ID-12LA

76
Summary

The concept of integration enables RFID components to be embedded within the

warehousing system’s infrastructure. This includes an integration of hardware and
software applications. This chapter describes the RFID-enabled inventory
management system integration approach and the types of radio frequency fields that
operate within RFID systems according to antenna signal. Under the integration of
RFID-enabled warehousing management system, customers place their orders online
through a web-based platform and the RFID-inventory management system can
automatically check its database in terms of availability of these ordered items. Key
components of the RFID system was identified and used for an experiment for
integration of the RFID-enabled inventory management system. Based on this, a pilot
test (described in the next chapter) of the RFID inventory management system was
carried out and it demonstrates the capability of interacting with the mechanical control
system. The RFID-data processing item selection algorithm was illustrated based on
the inventory selection process of ordered items to be dispatched by the warehouse
with respect to its criteria. The routing algorithm for a pusher to be given the priority to
travel to one of selected items was also illustrated

77
Chapter 7 A pilot study and experimental results

As discussed previously, one of the key elements of an RFID-enabled warehousing

system is its data processing mechanism, namely the warehouse management
system (WMS) that deals with in-store inventory data received by the RFID system.
This chapter provides detailed experimental results based on a pilot test. In order to
carry out a pilot test, a program, which is based on the developed mathematical model
to determine the estimated travel time for items to be dispatched from an AS/RR to a
specified collection point in order to determine the optimal solution (with the least travel
time), was developed using MATLAB (shown in Figure 44). A prototype of the
proposed RFID-based management system was built using two segments as follows.

a) Near field communication (NFC) using a microcontroller board on the Arduino

ATmega328.
b) Far field communication (FFC) using the Impinj Speedway Gen2 RFID.

The ATmega328 microcontroller is the main system controller and the ATmega328
compiler is written using C++ language. The Arduino near field antenna is operated in
125 KHz-13.56 MHz RFID standards. The compiler is a platform for the computer
program that converts source code (written in a computer language such as C++) to
accomplish tasks. Figure 35 shows a prototype of the RFID-inventory management
system which was built with its backbone using an Arduino interface which interacts
with the RFID management system. The ID-12 module was used to perform the task
of detecting and reading RFID tags and subsequently printing out RFID codes over a
USB port terminal interface (COM4) as shown in Figure 36.

78
7.1 Hardware and software of the RFID inventory management system

Figure 35: The prototype of the RFID-inventory management system using an

Arduino interface (Author’s completion).

Figure 36: Screen shot of an example of RFID codes generated through the Arduino
platform (Arduino open source software).

79
Figure 37 shows a screen shot of an example in the command window of information
collected by reading RFID tags in a MATLAB environment based on the following
rules:
(1) The location of an ordered item.
(2) The state of an ordered item.
(3) The identity of an ordered item at a particular location.
(4) The nearest object to the collection point.

The developed RFID-based management system can automatically detect and

identify a number of locations of the ordered item by displaying objects in rows and
columns on a storage rack. If there are several locations containing the same type of
the ordered item, based on the developed item selection algorithm, the system will
then issue a demand to dispatch the ordered item by assigning a priority to the
selected item based on the closest expiry date if applicable and the nearest location
of an item to a collection point.

80
 Total travel time
with constant
velocity

Figure 37: Screen shot of processed RFID data of an ordered item.

In order to determine the travel time of an ordered item as discussed previously in

section 5.4 in chapter 5, two scenarios are considered, i.e., in case I with a constant
velocity and case II with an acceleration and declaration. Figure 38 shows a
comparison in the experimental result of the total travel time obtained with the constant
velocity and acceleration/declaration, respectively. It indicates that the total travel time
obtained with acceleration/declaration is slightly higher than that using the constant
velocity without not much difference.

81
 Total travel time
with constant
velocity

Total travel time with

acceleration and declaration

Figure 38: Screen shot of the constant velocity vs acceleration/deceleration.

Figure 39 shows the hardware used for constructing the RFID-based warehouse
management system consisting of the following key elements:

1. An Impinj RFID reader product (Speedway R420). (Impinj company)

2. A UHFA Gen 2 RFID antenna.
3. Multiple UHF Gen 2 RFID tags.
4. Connect the reader to a PC (laptop) via Ethernet cable.

Figure 39: Testing of the RFID-based inventory management system.

82
7.2 Tests of RFID passive tags

An experiment was performed indoors in a room 7 m long and 5 m wide. A number of

RFID passive tags were placed on the ground throughout the room. The distance
between passive tags was 70-100 cm. Two pairs of far-field antenna were placed on
a desk table, which are fixed to the holder at a distance from the ground of 1 m, with
the tilt angle of 45 degrees toward the ground where RFID passive tags are. In order
to configure the MultiReader software platform (Impinj multireader software-configure
setting) as an example shown in Figure 40, the following steps were used Ensure the
latest version of multi-reader is installed/used.

1. Click on 'Configure Settings'.

2. Select 'Modes, RF, Power' menu - ensure desired antenna is checked
and that Rx Sensitivity Max is checked.
3. Select the 'Run Mode' menu
4. From the Run Mode drop-down box, select 'Margin Test'.
5. Ensure that the On Time is at least 10 seconds (10000 ms) for FCC
reader hops ~ every 200ms across 50 frequencies/channels so in 10 seconds
it will cover all available.

Figure 40: Screen shot of configure the MultiReader software platform (Impinj
multireader software-configure setting).

83
 6. Set the Start to desired value. Minimum power s +10dBm and max can
be up to 32.5dBm, depending on the region, model and power source. The
power can be set to increment in as little as 0.25dB steps (Shown in Figure 41).
7. Click 'Apply' - this will return you to the main screen.
8. Click 'Start Margin Test Run'.

Figure 41: Setting is applied correctly for the MultiReader software platform (Impinj
multireader software-configure setting).

The Impinj Speedway Gen2 RFID reader was used to perform the task of detecting
and reading RFID tags. RFID codes can be displayed on the MultiReader window as
shown in Figure 42 over an Ethernet port. This MultiReader software offers a platform
which allows a user to configure the RFID reader’s operational settings and tests.

84
 Figure 42: Screen shot of RFID codes generated through the MultiReader software
platform (Impinj multireader software).

7.3 RFID tags read rang test

The primary consideration in selecting RFID tags is to find the optimal balance
between size of the RFID tag and read range. The right read range is determined by
the physical distance between the assets and the RFID reader. The read range for the
selected RFID passive tags in this research is up to 9 m/30 ft. A test was carried out
to see how the RFID passive tags operate regarding the response time rate and the
ideal distance to place the RFID tag from the RFID antenna. Different read ranges
were performed for the RFID tags from 1 m to 9 m in order to make a comparison over
numbers of RFID tags used. The test showed that up to 4 m the read range gave
approximately the same time rate per second. This was considered to be the ideal
location to place an RFID tag. Nevertheless, in retailers, or any environment where
many assets are close together, a long read range can cause signal interference,
making it difficult to take inventory data on just one aisle or section. When assets are
in close proximity, a shorter read range works better.

7.4 RFID application development

The first step is to download the Octane SDK from Impinj intended to create the RFID
application using the programming language of C#/.NET under the Octane libraries.
Once the basic windows presentation foundation (WPF) solution was built in Visual

85
Studio, the Octane Library was imported ready for usage. Table 12 shows part of the
SDK programming code (C#) for processing RFID data in the MVS environment, which
is the window simulator shown in Figure 43 (the whole SDK programming is shown in
Appendix F).

Table 12: Part of the SDK programming codes for data processing of acquiring RFID
tags.
================================================================
When the user presses the button to start reading to get a new Rfid tag, it calls the
following function:
RfidTracker project: app\public\trackeditem.js getNewRfidTag()
This in turn calls app\public\services\repository.trackeditem.js getNextRfidTag()
RfidTrackerController.cs GetNextRfidTag()
This checks to see if we are using the simulation reader. If we are, we just return a
new fake tag id.
if (_repository.UseRfidSimulation)
{
rfidNewTagResponseToReturn = new RfidNewTagResponse()
{ ReturnCode = 0,
RfidTagValue = Guid.NewGuid().ToString(),
ErrorMessage = string.Empty };
}

Otherwise we call into the actual RfidReader code. This is in the

RfidTracker.Communications project in the RfidController.cs file. GetNewTag()

Running the Rfid Reader Console

{
USE [RfidTrackerWarehouse]
-- Get the last 10 records from the tag event table
SELECT TOP 10 * FROM [dbo].[RfidTagEvent] ORDER BY [DateAndTime] DESC
-- Get the last 10 records from the tracked item event table
SELECT TOP 10 * FROM [dbo].[TrackedItemEvent] ORDER BY [EventDate] DESC
-- Get the items that moved!
SELECT *
FROM [dbo].[TrackedItem]
WHERE [Id] IN (SELECT TOP 10 [TrackedItemId]
FROM [dbo].[TrackedItemEvent]
ORDER BY [EventDate] DESC
}

86
 Figure 43: MVS window simulator.
The GUI allows a matrix manipulation of item locations and it can interface with other
programs written in different languages such as C, C++, C#, Java etc. For instance,
as shown in Figure 44, a user can search an ordered item in the GUI by clicking on
the Search button and entering the item’s name (e.g., ‘‘Milk’’). The GUI office will then
indicate where the item is located in the map of storage racks. Once the item is
identified, information of this item can also be found by clicking on the Display button
with the displayed information including the identity of the ordered item, the state of
the ordered item (e.g., milk expiry date), the location of the ordered item (rows and
columns of storage racks), the number of the ordered items available and the nearest
item of the same type to a collection point. (In appendix D, it shows the GUI
programming codes).

87
 Items

Selected
item
Collection
point

Figure 44: Display of an optimal location of one of a selected item in the GUI.

7.5 MATLAB simulation results

The travel speed Vx is 210mm/s and the same travel speed was assumed for V y as
shown in Table 13, which explains the determination of an item with a longest travel
time in an AS/RR. Assuming there are three items which are located in storages
namely S1 (5, 3), S2 (4, 3) and S3 (5, 5), respectively. Let us 𝑉𝑥 refer to a speed of an
output conveyor and 𝑉𝑦 refer to a speed of a spiral conveyor. As stated previously, a
standby pusher has a default location at the centre (3, 3) of the AS/RR. Assuming
each grid, which represents a standard storage space of the AS/RR, has coordinating
dimensions in height and width of 1 x 1 meter.

To calculate the total travel time 𝑇𝑡𝑡_𝑐𝑖 for each selected item is obtained by:

𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝐶𝑜𝑙𝑜𝑢𝑚𝑛𝑠 𝑁𝑢𝑚𝑏𝑒𝑟 𝑜𝑓 𝑅𝑜𝑤𝑠

𝑇𝑡𝑡_𝑐𝑖 = 𝑇ℎ + 𝑇𝑣 = +
𝑉𝑥 𝑉𝑦

88
Where 𝑇ℎ 𝑎𝑛𝑑 𝑇𝑣 refer to the horizontal and vertical travel time of a selected item,
respectively.

Table 13: Determination of an item with a longest travel time in an AS/RR to the
collection point.

Vx S1
S3
(5,1) (5,2) (5,3) (5,4) (5,5)

(4,1) (4,2) (4,3) (4,4) (4,5)

(3,1) (3,2) (3,3) (3,4) (3,5) S2

Vy
(2,1) (2,2) (2,3) (2,4) (2,5)

(1,1) (1,2) (1,3) (1,4) (1,5)

1m
1m
To the collection point

3 ∗ 1000 5 ∗ 1000
𝑇𝑡𝑡_𝑐𝑖 (𝑆1 ) = + = 38.09 𝑠𝑒𝑐
210 210

3 ∗ 1000 4 ∗ 1000
𝑇𝑡𝑡_𝑐𝑖 (𝑆2 ) = + = 33.32 𝑠𝑒𝑐
210 210

5 ∗ 1000 5 ∗ 1000
𝑇𝑡𝑡_𝑐𝑖 (𝑆3 ) = + = 47.6 𝑠𝑒𝑐
210 210

Based on these calculated results, the RFID-inventory management system can

determine the optimal route for the pusher to travel to select an item in a sequence as
shown in Figure 45.

Figure 45 shows the MATLAB simulation result which determines an item with a
longest travel time from the selected items to be pushed in a sequence onto the output
conveyor by a pusher from an AS/RR, i.e., the item with a longest travel time is given
a priority to be pushed by the pusher to travel to a specified collection point until the
last selected items with the least travel time arrives to the same destination. This

89
allows a minimum waiting time for packers to receive all the selected items at the
collection point. As shown in Table 13 as an example, the result indicates that the
selected item at S3 (5, 5) has a longest travel time of 47.6 seconds and this item should
be given the highest priority to be pushed onto an output conveyor travelling to the
specified collection point. The selected item at S2 (4, 3), however, has a least travel
time of 33.32 seconds and this item (i.e., the last remaining item shown in Figure 46)
should be given the least priority to be pushed onto an output conveyor travelling to
the specified collection point. The selected item at S1 (5, 3) has a travel time of 38.09
seconds and therefore it should be the second item to be pushed onto an output
conveyor towards the same destination. However, if there are two ordered items which
have the same travel time, the RFID-based management system will be set as a
default to select the ordered item that is close to the pusher. The result was generated
by the RFID-inventory management system according to the pre-defined selection
rules and the system subsequently issues a priority to be given to the selected item.
The mechanical control system then initialise a demand to push the selected item onto
an output conveyor and this item will be transported along the RFID-guided route to
the collection point (Other result can be found in Appendix I).

Figure 45: Determination of a longest travel time among ordered items.

90
Figure 46 shows the accumulating results of travel times for each selected item which
travels through the five different sections along the RFID-guided route from the
moment when the pusher is activated to the moment these selected items arrive at a
collection point. It shows the travel time Tij in minutes between the activated pusher i
and the selected item j. Tjk refers to a travel time from the moment when the pusher i
starts to push a selected item j to the moment that the selected item j is pushed onto
an output conveyor k. Tkl refers to a travel time of a selected item j which travels at a
constant speed along an output conveyor k to the entrance point of a spiral conveyor
l. Tlm refers the travel time from the top level l to the bottom level m through a powered
spiral output conveyor. Finally, Tmn refers to the travel time between the end of the
spiral conveyor m and the collection point n. As shown Figure 46, item 3 has the
longest travel time Tij of 15.6 seconds, whereas item 2 has the least travel time of 3.23
seconds as the default location of the pusher is closer to item 2. All the selected items
have the same travel time Tjk as the storage space for each item has a standard height
and width of 1 x 1 meter. Either of items 2 or 3 has the same travel time Tkl of 12
seconds because these items are located at the same column in the AS/RR. Overall,
item 3 has the longest accumulative travel time of 47.6 seconds and item 2 has the
least travel time of 33.32 seconds. Thus, the developed programing model of the
RFID-enabled inventory management system can generate the priority list of selected
items to travel in a sequence from the first item with a longest travel time to the last
item with a least travel time, accordingly.

91
Figure 46: The results of selected items for five different stages with accumulating
travel times.

92
Summary

This chapter describes the implementation of a pilot test based on an integration of

hardware and software towards the development of an RFID-based management
system and the results were obtained and analysed in a MATLAB environment. Also,
the prototype of the proposed RFID-based management system was built and tested
using near field communication (NFC) and far field communication (FFC), respectively.
The developed RFID-based management system can automatically detect and
identify locations of ordered items by displaying objects on GUI. Experimental results
demonstrate the applicability of the developed optimisation algorithm implemented
into the RFID-based inventory management system through the pilot test. It shows
that the RFID-inventory management system can determine the optimal route for the
pusher to travel in a sequence during the ordering process that gives a minimal overall
travel time for all the ordered items to a specified collection point.

93
Chapter 8 Multi-Objective optimisation for the RFID-enabled automated
warehousing system

The objective function involves an optimisation of non-linear and multi-objective

variables in such as (i) number of storage and retrieval (S/R) machines, (ii) dimensions
of the SR and (iii) costs of building and operating warehouses and so on. One of the
early studies in optimising the design of warehouses was reported by Basan et al.
(1980). The study was aimed at seeking optimum dimensions of the warehouse
corresponding to warehouse volume under storage strategies. Similarly, Ashayeri et
al. (1985) illustrated a design model of the AS/RS that enables the determination of
the influential parameters. Bafna et al. (1972) & Perry et al. (1983) presented a study
using a combination of the analytical model the discrete event simulation when
designing the warehouse. Rosenblatt et al. (1984) offered the design of warehouses
in regard to the influence of the storage policy by considering the following costs: (i)
cost of building the warehouse, (ii) costs of buying storage equipment, (iii) costs arising
from overloading the warehousing system (temporary shortage of the storage space)
and (iv) costs associated with one of particular storage policies. Although some
researchers reported studies in optimisation of travel time, cost, service level and
capacity utilisation individually based on requirements of warehouse design specified,
these studies do not consider the optimisation of all the above parameters using the
multi-objective approach providing a trade-off solution particularly in automated
warehouses based on the literature review.

In practice, many engineering design problems are multi-objective problems as these

often involve more than one design goal to be optimised. These design goals
potentially impose conflicting requirements on the technical and economic
performances of a given system design. To determine a trade-off solution in alternative
design options between these conflicting design objectives, an optimisation problem
with multiple objectives often need to be resolved. To facilitate the exploration of
different multi-objective formulations, Pareto optimality concepts can be incorporated
into optimisation algorithms as a decision maker. Multi-objective optimisation refers to
an optimisation of multiple decision making objectives concurrently. These objectives
are possibly conflicting and competing. In a multi-objective problem, however, it is
often impossible to obtain a single optimal but a trade-off solution among a number of
objectives, since there is a contradictory among antagonist objectives. For further

94
details on multi-objective optimisation, it can be referred to publications by Coello et
al., 2007; Miettinen, 1998; Collette and Siarry, 2011; and Rangaiah and Bonilla-
Petriciolet, 2013.

Theoretically, a compact multi-objective optimisation can be formulated as follows:

max/ min O( x)  (O1 ( x), O2 ( x),..., O f ( x))T (2.1)

subject to

gi ( x)  0, i  1, 2,3,..., a hj( x)  0, j  1, 2,3,..., b. (2.2)

Where

O number of objective functions

a and b number of constraints

x  B z the decision variable vector

z number of independent variable xi

In the multi-objective optimisation, unlike the single objective optimisation, there is no

mono-dominant solution but a set of non-dominant solutions called Pareto (or non-
dominant, non-inferior) solutions. Pareto solutions denote a set of compromised
solutions between the conflicting objectives. Ahmadi et al. (2015) formulated a multi-
objective optimisation electric model to integrate the generation of thermal units
considering heat and power dispatch. The objective functions of the proposed multi-
objective framework comprise simultaneous minimisation of cost and thermal units’
emission as well as maximising heat generation. Mohammed et al. (2017) presented
a development of a product distribution planner for a three-echelon green meat supply
chain design in terms of issues including numbers and locations of facilities that should
be opened in association with the product quantity flows. These issues were
formulated into a fuzzy multi-objective programming model (FMOPM) with an aim to
minimise the total cost of transportation and implementation, the amount of CO2
emissions in transportation and the distribution time of products from farms to abattoirs
and from abattoirs to retailers, and maximise the average delivery rate in satisfying
product quantity as requested by abattoirs and retailers. Ramesh et al. (2013)
proposed a modified non-dominated sorting genetic algorithm-II (MNSGA-II) which
was applied to a multi-objective reactive power planning problem by incorporating the

95
concept of Dynamic Crowding Distance (DCD) in NSGA-II algorithm. Three conflicting
objectives combined operating and VAR allocation cost minimisation, bus voltage
profile improvement and voltage stability enhancement were considered. The IEEE
30-bus system, the 69-bus system and the IEEE 118-bus system were also
considered. By using conventional weighted sum method with Covariance Matrix
Adapted Evolution Strategy (CMA-ES), a reference Pareto-front was generated.
Pishvaee et al. (2012) proposed a multi-objective fuzzy mathematical programming
model for designing an environmental supply chain under inherent uncertainty of input
data in such problem. The proposed model was able to consider the minimisation of
multiple environmental impacts beside the traditional cost minimisation objective to
make a fair balance between them. A life cycle assessment-based (LCA-based)
method was applied to assess and quantify the environmental impact of different
options for supply chain network configuration. Also, to optimise the proposed multi-
objective fuzzy optimisation model, an interactive fuzzy solution approach was
developed.

This chapter presents a development of a mathematical model using the multi-

objective approach as an aid for optimising the design of a proposed RFID-
enabled automated warehousing system. The developed model used for maximising
the warehouse capacity utilisation and service level in terms of satisfying all
demands of dispatching products from the warehouse, minimising the travel distance
of products from a storage rack to a collection point and the total cost required
for implementing RFID-related facilities into the proposed warehousing system. This
chapter also includes an investigation into configuration of the proposed
warehousing system focusing on the optimal number of racks and collection
points that need be established.

8.1 Problem definition and model description

One of the main issues to be addressed is to optimise the design of the proposed
RFID-enabled automated warehouse including the allocation of the optimum number
of racks and collection points towards four objective functions (as illustrated in Figure
16 in chapter 4): (1) minimisation of total cost of implementing the warehouse, (2)
maximisation of capacity utilisation of the warehouse, (3) maximisation of service level
in terms of satisfying all demands of dispatching products from the warehouse and (4)

96
minimisation of travel distance of products from racks to collection points. The
following sets, parameters and decision variables were used in the formulation of the
model:

Sets:

I set of racks i  I
J set of collection points
j J
K set of departure points
k K

Given parameters:

Cir fixed cost required for establishing a RFID-enabled rack i

Cic fixed cost required for establishing a collection point j

Cit unit RFID tag cost per item at rack i

CijT unit transportation (T) cost per meter from collection point i to
departure point k
C lj unit labour cost per hour at collection point j

R lj working rate (items) per labourer ( l ) at collection point j

d jk travel distance (meter) from collection point j to departure

point k
W transportation capacity (units) per forklift
Si maximum supply capacity (units) of rack i
Sj maximum supply capacity (units) of collection point j
Dj demand (in units) of collection point j
tijd average travel distance (in meters) per item from rack i to
cv collection point j

Decision variables
qij quantity of units ordered from rack i to collection point j
q jk quantity of units dispatched from collection point j to

departure point k
xj required number of labourers at collection point j

yi 1: if rack i is required
0: otherwise

97
 yj 1: if collection point j is required
0: otherwise

The four objectives, which include a minimisation of total cost, a maximisation of

capacity utilisation, maximisation of service level and minimisation of travel distance,
are formulated as follows:
Objective function (1)

In this case, the total cost of establishing the RFID-enabled automated warehouse
includes costs of establishing RFID-enabled racks, collection points, RFID tag,
transportation of products and labours. Thus, minimisation of the total cost can be
expressed below:

Min F1   Cir yi   C cj y j    Cit qij   CijT q jk / W f  d jk (1)

iI jJ iI j j jJ kK

 C x j
i
l

iI

Objective function (2)

Maximisation of capacity (C) utilisation is expressed as follows:

1 (2)
 2 2
  Cactual    Cused   
Max F2   
 iI

 i 


qij  q ij
Where Cactual   and Cused 
iI jJ

jJ Si S iI
i

Objective function (3)

To ensure the satisfaction of all demands of products delivered by the warehouse, this
can be achieved by maximisation of service level, which is given by:
qij (3)
Max F3  
iI jJ Di

Objective function (4)

The travel distance of an in-store item from its location of a storage rack to a collection
point can be minimised as follows:

98
 (4)
Min F4   t ijd qij
iI jJ

There are a number of constraints which are included in the optimisation. The
constraints are given as follows:

q
iI
ij  Si yi  j J (5)

q
jJ
jk  Sj yj k  K (6)

q
iI
ij  Dj
 j J
(7)

D j  q jk (8)
kK  j J

q
jJ
ij  x j R lj  iI (9)

qij , q jk  0, i, j, k ; (10)

yi , y j 0,1 , i, j; (11)

Equations 5 and 6 refer to the flow balance of a product travelling from a storage rack
to a collection point and from a collection points to a departure point. Equations 7 and
8 refer to demands to be satisfied. Equation (9) determines the required number of
labours at a collection point. Equations (10) and (11) limit the decision variables to
binary and non-negative.

8.2 The optimisation methodology

Several approaches were reported in the literature to solve multi-objective optimisation

problems. In this work, a fuzzy solution approach was proposed to transform the multi-
objective model into a single-objective model which is formulated by considering each
objective individually by minimising the secularised difference between value of each
objective and its optimal value. Undesired deviations were used for being subtracted
from the single objective function to obtain more accurate objective values.

A. Solution procedures

There are a number of methods proposed in the previous literature to deal with
possibilistic programming models (e.g. Lai et al., (1992), Jimenaz et al., (2007) &
Liang, (2006). The Jimenez et al. (2007) method is formed based on the strong

99
mathematical concepts, i.e., expected interval and expected value of fuzzy numbers,
also this method is computationally efficient to solve fuzzy linear problems as it can
preserve its linearity and does not increase the number of objective functions and
inequality constraints.

To solve the optimisation problem based on the developed multi-objective model, the
solution procedures are expressed as follows:

1) Convert the developed model into an equivalent crisp model using Jiménez
method (Jiménez et al., 2000). Accordingly, the equivalent crisp model can
be formulated as follows.

4 Cipr 4 C cjp 4 Cipt (12)

Min F    yi    yj    qij
iI p 1 4 jJ p 1 4 iI j j p 1 4
T 4 Cl
4 Cijp
   q jk / W f  d jk    ip x j
jJ kK p 1 4 iI p 1 4
1
(13)
 2 2
  Cactual    Cused   
Max F2   
 iI

 i 


4 qij (14)
Max F3   4
iI jJ p 1 Dip

(15)
Min F4   t ijd qij
iI jJ

Subject to:

q
iI
ij  Si yi  j J (16)

q
jJ
jk  Sj yj k  K (17)

 D j1  D j 2    D  Dj4 (18)
q
iI
ij 
2 2
 1   j 3
 2 2  j J
 D j1  D j 2    D  Dj4 (19)
.  1   j 3  q jk
2 2  2 2 kK  j J

100
  x j1  x j 2    x j3  x j4 l (20)
q
jJ
ij  xj
2
.
2
 1  
 2 2
Rj  iI

qij , q jk  0, i, j, k ; (21)

yi , y j 0,1 , i, j; (22)

2) Find the upper and lower bound (U, L) solution for each objective function.
This can be obtained as follows for
Upper bound solutions:

Max F1 (U1 )   Cir yi   C cj y j    Cit qij  CijT  q jk / W f  d jk (23)

iI jJ iI j j jJ kK

 Cil x j
iI
1
(24)
 2 2
  Cactual    Cused   
Max F2 (U 2 )    
 iI

 i 

qij
Max F3 (U 3 )   (25)
iI jJ Di

Max F4 (U 4 )   t ijd qij (26)

iI jJ

Lower bound solutions:

Min F1 ( L1 )   Cir yi   C cj y j    Cit qij  CijT q jk / W f  d jk (27)

iI jJ iI j j jJ kK

 C x j i
l

iI
1
(28)
 
 Cactual    Cused  
2 2

Min F2 ( L2 )     

 iI

i 

qij
Min F3 ( L3 )   (29)
iI jJ Di

Min F4 ( L4 )   t ijd qij (30)

iI jJ

3) Find the respective satisfaction degree µ(xi) for each objective function as
follows:

101
 1 if F1 ( x )  U1 (31)

 F ( x)  L1
1 ( F1 ( x))   1 if L1  F1 ( x)  U1
 U1  L1

0 if F1 ( x )  L1
1 if F2 ( x)  U 2 (32)

 F ( x)  L2
2 ( F2 ( x))   2 if L2  F2 ( x)  U 2
 U 2  L2

0 if F2 ( x)  L2
1 if F3 ( x )  U 3 (33)

 F ( x)  L3
3 ( F3 ( x))   3 if L3  F3 ( x)  U 3
 U 3  L3

0 if F3 ( x )  L3
1 if F4 ( x)  U 4 (34)

 F ( x)  L4
4 ( F4 ( x))   4 if L4  F4 ( x)  U 4
 U 4  L4

0 if F4 ( x)  L4

4) Transform the crisp model obtained from step (1) to a single objective function
using the developed solution method. The developed solution method function
(F) is formulated as follows:

 4 4  4 (35)
Min F   n  f ( x)   Fd ,  n 1
 n 1 f 1  n 1

Set n*  n Fn , then



Fn  Fn
Fd   F  2* F2  3* F3 4* F4
*
1 1
(36)
1 F1 2 F2 3 F3 4 F4
 F1  FZ  F3  F4
F1  F1 F2  F2 F3  F3 F4  F4
(37)
Based on the aforementioned procedures, the developed method’s objective
function can be written as follows.

  F  F  F  F 
Min F  11  2 2  3 3  4 4     1 1 F1   2 2 F2   3 3 F3   4 4 F4 
 F1  F1 F2  F2 F3  F3 F4  F4 
 
Subject to equations (5)-(11).

102
 5) Vary the weight combination set (  ) consistently for the four objectives to reveal
Pareto-optimal solutions. Usually, the weight combination set is allocated by
decision makers based on the importance of each objective.
6) Select the best Pareto-optimal solution using the proposed decision making
method.

B. The decision making method: TOPSIS

The next step after revealing the Pareto solutions is to determine a best trade-off
solution, the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS)
method was employed for revealing the best trade-off solution. TOPSIS is currently
one of the most popular multiple criteria decision making (MCDM) methods in a
number of application areas. This method was first developed by Hwang and Yoon
(1981) for solving MCDM problems with the basic principle to choose the alternative
which has the shortest distance from the positive ideal solution (best) and the farthest
distance from negative-ideal solution (worst). In this research work, the TOPSIS was
employed for revealing the best trade-off solution. This approach selects not merely
the closest solution to the ideal solution, but also the farthest from the negative ideal
solution (Ramesh et al., 2013). TOPSIS was implemented as follows:

 
Assume PR- PR op o=1,2,...,x (number of pareto solutions); p =1,2,...,y (number of objectives)

where PR is a x * y decision matrix, which represents the performance rating of

alternative Pareto solutions with respect to the objective function values. Thus, the
normalized selection formula is presented as follows:

PRop (38)
NPR  o

 PR
p 1
ap

The value of a decision can be measured by the entropy value as:

1 x (39)
Ep   PRop ln( PRop )
ln x o1

The degree of divergence Dp of the average intrinsic information contained for p = 1,

2, 3, 4 can be calculated as:

103
 Dp  1  E p (40)

The weight for each objective function value is given by:

Dp (41)
wp  y

D
k 1
k

Thus, the objective weighted normalized value is given by:

vop  wo PRop (42)

The positive ideal solution AT and the negative ideal solution AT are taken to generate
an overall performance matrix for each Pareto solution. These values can be
expressed as:
AT   (max(vo1 ) max(vo 2 ) max(voy ))  (v1 , v2 ,..., v y ) (43)

AT   (min(vo1 ) min(vo 2 ) min(voy ))  (v1 , v2 ,..., v y )

Distance between alternative solutions can be measured by the n-dimensional
Euclidean distance. The separation of each alternative from the ideal solution is given
as

 y  (44)
Dp   (v po  vo ) 2  , p  1, 2,..., x
 o 1 
 y  (45)
Dp   (v po  vo ) 2  , p  1, 2,..., x
 o 1 

The relative closeness to the ideal solution of alternative solutions with respect to
objective function values is expressed as follows:
Dp (46)
rc p   
, p  1, 2,..., x
D D p p

Where Dp  0 and Dp  0 , then, clearly, rc p  1,0 . Select the trade-off solution with
 

maximum rcp or list the obtained solution in descending order based on rcp.

104
8.3 Application and evaluation

In this section, a case study was used to investigate the applicability of the developed
optimisation model and the performance of the proposed optimisation methodology.
Table 14 shows the parameters used for application. The solver for the developed
multi-objective model was LINGO11.

Table 14: Application data ranges.

= 12 I Si = 25K-35K
J = 15 S = 20-29K
j

K = 2

C = 6.5-9 (GBP)
l
j d jk = 20-45
C = 60-90
r
i t d
ij
= 40-75
(KGBP)
C = 0.25 (GBP)
t
i Dj = 150-360 K
C = 0.4 (GBP)
T
jk d = 20-45
jk

R = 100 l
j C = 15-18 c
j

(KGBP)
W = 48

8.3.1 Results and discussions

This section presents the computational results obtained based on the developed
multi-objective model using the proposed integrated fuzzy solution approach for the
problem previously defined.

The solution steps of the developed model are described as follows:

1) Obtain the upper and lower value for each objective function by solving them
individually. The results are ({UF, LF}) = ({504, 1,230}, {0.66, 0.94}, {0.85, 0.99},
{400, 2,310}).
2) Find the respective satisfaction degree µ(xi) for each objective function. The
satisfaction degrees are reported in Table 15.
3) Solve the crisp model as a single objective model using the developed solution
method by an assignment of different combination of weight values with respect
to the feasibility of each weight pair that is denoted by λ.
4) Select the best solution using TOPSIS, the determined score values of Pareto-
optimal solutions are reported in Table 16.

105
 Table 17 shows the obtained set of Pareto-optimal solutions based on the four
objective functions. It also shows the number of racks and collection points that should
be established. For instance, solution 1, which is obtained by an assignment of
1  1,2  0,3 = 0 and 4  0 , has a minimum total cost of 504 thousand GBP, a maximum

capacity utilisation of 66%, a maximum service level of 85% and a minimum travel
distance of 400 km of all products. This solution offers six racks and nine collection
points. Figure 47 illustrates the further comparison among the solutions in response
to the four objective functions. It can be observed in Figure 47 that the Pareto optimal
method cannot produce a better solution in one objection function without worsening
its performance in the other objective functions. After obtaining a set of Pareto-optimal
solutions, decision makers may determine a solution depending on their preferences
or using a decision making algorithm. Note that the decision maker usually selects an
acceptable solution belonging to the Pareto front. Identifying a set of Pareto optimal
solutions is thus a key point for the decision maker’s selection of a compromise
solution satisfying all the objectives as better as possible (Chiandussi et al., 2012).
The multi-objective optimisation problem (also called multi-criteria optimisation, multi-
performance or vector optimisation problem) can then be defined as the problem of
finding (Osyczka et al., 1985) ‘‘a vector of decision variables which satisfies constraints
and optimises a vector function whose elements represent the objective functions.
These functions form a mathematical description of performance criteria which are
usually in conflicting with each other. Hence, the term ‘optimise’ means finding such a
solution which give the values of all the objective functions acceptable to the decision
maker’’. In this research work, TOPSIS was employed to select the best solution. As
shown in Table 16, solution 6 is the best solution as its rcp score is the highest 0.279
with the total cost of 795 thousand GBP, 84% capacity utilisation, 93%-check again
these figures are fo Solution 4 service level and travel distance 1198 km for all ordered
products. Also, it requires an establishment of nine racks to supply products to eleven
collection points.
Table 15: Satisfaction degrees for objective functions.
µ(x1) 0.95 0.93 0.9 0.88 0.82 0.77 0.74 0.7 0.688 0.646 0.622 0.6 0.57 0.55

µ(x2) 0.7 0.73 0.75 0.77 0.79 0.8 0.84 0.85 0.88 0.90 0.92 0.97 0.98 0.99

µ(x3) 0.73 0.74 0.77 0.78 0.79 0.82 0.84 0.85 0.89 0.91 0.92 0.97 0.97 0.99

µ(x4) 0.97 0.96 0.95 0.93 93 0.91 0.9 0.88 0.86 0.84 0.81 0.79 0.76 0.72

106
 Table 16: Score values of Pareto-optimal solution using TOPSIS.
Solutio 1 2 3 4 5 6 7 8

Score 0.245 0.234 0.266 0.245 0.2544 0.279 0.273

9 10 11 12 13 14

0.244 0.249 0.256 0.249 0.271 -

107
 Table 17: Computational results.

350 non-zero elements, 59 constraints, 111

total variables, 59 integer variables
Feasibility # Min Max Max Min Open Open
level (F1) (F2) (F3) (F4) racks collection
(KGBP) (%) (%) (Km) points
0.3 1 504 0.66 0.85 400 6 9

0.35 2 564 0.68 0.85 680 6 9

0.4 3 595 0.71 0.87 873 7 8

0.45 4 678 0.78 0.89 932 7 8

5 5 720 0.80 0.92 1011 8 9

0.55 6 795 0.84 0.93 1198 9 11

0.6 7 844 0.86 0.94 1230 10 11

0.65 8 894 0.89 0.95 1342 10 12

0.7 9 940 0.90 0.96 1555 11 13

0.8 10 978 0.92 0.97 1701 11 13

0.85 11 998 0.93 0.98 1890 12 13

0.9 12 1064 0.93 0.98 2030 12 13

0.95 13 1099 0.94 0.99 2197 13 14

1 14 1120 0.94 0.99 2280 13 14

108
 0.95 1
Capacity utilization (%)

0.9 0.98

Service level (%)

0.96
0.85
0.94
0.8 0.92
0.75 0.9
0.7 0.88
0.86
0.65 0.84
0.6 0.82
504 595 678 795 894 978 1064 504 595 678 795 894 978 1064
Total cost (Thousand GBP) Total cost (Thousand GBP)
Total cost vs Capacity utilization Total cost vs Service level

2130 0.95
1930
Capacity utilization (%)
Travel distance (Km)

0.9
1730
1530 0.85
1330 0.8
1130 0.75
930
0.7
730
530 0.65
330 0.6
504 595 678 795 894 978 1064 0.85 0.85 0.89 0.955 0.955 0.97 0.985
Total cost (Thousand GBP) Service level (%)
Total cost vs Travel distance Service level vs Capacity utilization

Figure 47: Comparison among the obtained objective function values.

Figure 2. Comparison among the obtained objective function values.

109
Summary

This chapter presents a study in optimisation of the proposed automated warehouse

using the multi-objective optimisation approach. A multi-objective model was
developed and used for (1) optimising the design the proposed RFID-enabled
warehousing system in determining an optimum number of storage racks and
collection points that need be established for the warehouse, and (2) obtaining trade-
off decisions by measuring four conflicted objectives: minimisation of the total cost,
maximisation of capacity utilisation, maximisation of service level in terms of satisfying
all demands of dispatching products from the warehouse and minimisation of travel
distance in the warehouse. To obtain Pareto-optimal solutions based on the developed
model, an integrated fuzzy solution approach was proposed. Subsequently, a decision
making method was applied to select the best Pareto-optimal solution. A case study
was applied and the results demonstrate the applicability and validation of the
developed model.

110
Chapter 9 Cost analysis, benefits and limitations – a real case study

The University of Portsmouth library converted their current EM barcode Library self-
service and security solutions to a RFID solution in 2012. The implementation of the
RFID system enables the University of Portsmouth library service to improve
dramatically and meet their self-service aspiration enabling higher levels of throughput
to deliver improved customer service levels and return on investment.

9.1 Cost estimates

Table 18 provides an RFID implementation cost estimates for the library at University
of Portsmouth by the Bibliotheca Company. It lists each component description,
quantity needed and unit price.

Table 18: Self-service and security solution cost.

Description Qty. Unit price

smartserve™ 400 - barcode user ID included 5 £6,995.00
smartserve™ 400 payment frame 5 £795.00
*RRP £795.00/ unit incentive 5 x free of charge
smartserve™ 400 Chip & PIN 5 £595.00
*RRP £595.00/ unit incentive 5 x free of charge
smartadmin™ Software 5 £99.00
remote monitoring software per kiosk per year
smartstation™ 100 5 £750.00
*RRP £825.00/ unit incentive 5 x free of charge
smartstation™ 100 5 £750.00
*RRP £825.00/unit Incentive price @ £750.00/unit
smartlabel™ 200 430,000 £0.12p
*RRP £0.14p/ label incentive price £0.12p/ label
smartgate™ 300 single aisle 1 £4,990.00
smartgate™ 300 single aisle ramp main exit 1 £550.00
smartgate™ manager server licence 1 £500.00
smartgate™ manager client licence 2 £75.00
main exit gate & gates for refurbishment
smartgate™ 300 dual aisle 1 £4,990.00
*RRP £7,495.00 incentive £4,990.00 for refurbishment
NO RAMP quoted
Smartstock™ 300 3 £3,495.00
*RRP £3,795.00/unit incentive £3,495.00/ unit
Site Survey 1 £400.00
Installation/Implementation days 4 £550.00
Delivery Freight at cost 1 £1900.00
Total Exc VAT £116,985.00

111
9.1.1 Cost analysis for the RFID system vs barcodes

Table 19 lists the total implementation cost and running cost for both barcoding and
RFID technologies at the library of the University of Portsmouth. These figures were
provided by Ken (Project manager at University of Portsmouth). The barcode running
costs were estimated for each year and was terminated end in 2011, and the RFID
running costs were estimated for each year starting at 2012 and its still currently
ongoing project.

Table 19: Costs for RFID and barcodes at the university of Portsmouth library.

Type RFID Barcodes

Starting cost £116,985 £60,255
Running cost £6,039 £8,350 per annum

In other aspects, time saving, fast accessing of books and elimination of manual errors
are the main benefits of the RFID reported at the library. RFID can also be used in
library as anti-theft systems. There are some potential savings using RFID because a
single RFID tag can serve many different functions, compared to a barcode. The
library saves some time in processing new items because each item is attached with
an RFID tag. However, financial savings were not the goal of their RFID
implementation for the library at the University of Portsmouth, although it can be used
for that purpose through staff savings. Decision makers were looking for it to deliver
service transformation to be self-service. The implementation of the RFID system
enables the University of Portsmouth library service to improve dramatically and meet
their self-service aspiration enabling higher levels of throughput to deliver improved
customer service levels and return on investment.

112
9.1.2 The interview with project manager

An Interview was conducted with the project manager working at a strategic level
within the library at University of Portsmouth. The interview was free flowing using a
semi-structured approach, which allows the questioning and discussion greater
flexibility, offering the ability to adapt questions throughout the interview. Interview
focused on RFID deployment discovering the advantages and disadvantages of both
barcode and RFID methods so that the data can be effectively analysed.

9.1.3 The interview method includes the questions in such as:

1. General Information

a. What is the name of the organisation you work for?

b. What is your position in the library?
c. Which technology does your organisation use, barcodes, RFID or both?
2. How does the RFID system improve the efficiency within the library?
3. Books (stock) loss?
4. Real time inventory?
5. Has the performance of the library operations been improved?

A significant factor using RFID can offer fully automated reading of all tagged items,
whereas barcodes generally need some form of manual manipulation and scanning.
Also, application of RFID is less prone to human error. RFID allowed University of
Portsmouth library to achieve over 95% performance efficiency of issuing books. In
handling returns, an automated sorter pre-sorts books into broad areas reducing the
time taken to return books back to the shelves from a purely manual process. RFID
provides a more robust security system so that stock loss has been reduced due to
such as stealing books, miss-shelving or being hidden as the handheld scanner can
be used to locate items much more effectively. By using RFID system at the library,
human errors have significantly been reduced. However, errors do occur in the
incorrect programming of the RFID tags, although RFID provides a much more
accurate record of the items in the Library and the ability to locate individual items
much more effectively. Nevertheless, the library does not use the RFID system for real
time inventory as this does not give sufficient operational advantage to justify the
investment necessary in smart shelving. Implementation of the RFID system was the

113
starting point for a range of process change within the Library in which has been aimed
at improving efficiency and providing a better service for their customers.

The interview was also carried out with the project manager who was in charge of this
project into the following issues:

 How is the error comparable between RFID and Barcodes system?

 How efficient tracking system when items such as books were remotely placed?
 How far is the maximum distance to track?
 What happens beyond the library?

RFID has enabled the University library to reduce manual handling activities
associated with the issue of library stock by approximately 95% and with returns by
approximately 50% freeing up staff time to provide additional services such as the
welcome desk, roaming support and online chat support. The disparity in manual
handling reduction between ‘’issue’’ and ‘’return’’ is due to the nature of automatic
sortation which still requires items to be retrieved from a bin box manually. However,
this process is significantly takes less time intensive than sorting every individual item
by hand. Machine sorting is 100% reliable whereas human sorting with an experienced
member of staff is at best 95% reliable with the number of items being handled. This
reduces the human error count for items being miss-shelved considerably.

The use of RFID handheld readers (scanning process) allows the University library
staff to locate items quicker that have been miss-shelved, by either customer directly
or by students, or items that have been returned but were not removed from customers
account in the library management software. These processes used to be very labour-
intensive work as they involved members of staff visually scanning each item on the
shelf. Whenever a specific item is needed, the RFID handheld can be used to search
an item at the shelf for the presence of an RFID tag that is associated with a specific
item. With most ultra-high frequency (UHF) tags that work well for item tracking, read
ranges of 5-8 feet might be no problem. When doing activities within library manually
it however takes around four weeks to check each shelf using a team of 5-10 staff
members. By using the RFID handheld reader, tasks can be achieved/undertaken
within approximately a day or two days at a maximum by only one member of staff
representing a significant increase in operational efficiency which again allowed staff

114
to be deployed to activities with greater value for library customers. It was reported by
the project manager that without the introduction of RFID we might not have been able
to introduce these new services and might have lost out on significant customer
service advances or been able to achieve the customer service excellence
accreditation’’.

115
Summary

This chapter presents a study in cost analysis, benefits and limitations observed based
on a real case study regarding the implementation of the RFID technology into the
library at the University of Portsmouth. A cost comparative study between the barcode
system and the RFID system was provided. This was followed by the consideration of
labour, human errors, and efficiency by looking into the following issues: How is the
error comparable between the RFID system and the barcodes system at the library of
the University; how efficient the RFID tracking system is when items such as books
were remotely place? How far is the maximum distance to track? what happens
beyond the library? and other issues based on an interview with the project manager
at the university of Portsmouth library.

116
Chapter 10 Summary, discussions, conclusions and future work

This thesis presents an investigation into an RFID-inventory management system for

future generation automated warehousing systems through developments of a
number of design methodologies and integration approaches using fast-growing IT
technologies such as radio frequency identification (RFID) tags, wireless sensors and
communication networks. The thesis also presents a study in development of a using
multi-objective optimisation model for obtaining trade-off decisions based on a number
of conflicting objectives: minimisation of the total cost, maximisation of capacity
utilisation, maximisation of service level and minimisation of travel distance within the
proposed warehousing system.

Chapter 1 provides an introduction, background and research problems that motivated

this work to be investigated. Chapter 2 presents the findings about warehouse
operations in retail industry through a survey to identify some companies’ current state
of warehouses and operations in their organisation, and expectations in future
generation warehouses. Chapter 3 presents studies through the literature review in
conventional warehousing systems (manual/automated) and RFID-based automated
warehouse systems and other related research work and applications in logics and
supply chain sectors. The literature review is particularly focused on research work in
warehouse inventory management systems which are discussed in order to identify
the research issues that need to be addressed. Besides, advanced planning and
scheduling methods were discussed for production planning and scheduling systems
like MRP/ERP systems and lean management method. The Internet of Things (IoT),
was also discussed within this chapter. The chapter also examines RFID and sensor
network technologies including the background of the RFID and its application mainly
for supply chain and logistics sectors and review the applications that have been
established by researchers. A comparison between barcodes and RFID methods was
provided within this chapter. Travel time covers a substantial part of picking processes
in warehouses in which the item can be dispatched to travel from the storage rack to
the collection point, the relationship between the operation and cost in the warehouse
was also explained. Chapter 4 describes and discusses the key components of the
proposed RFID-based warehousing management system. This includes RFID-related
components, their mechanisms and features, and applications and its implementation.
In chapter 5, a material-handling solution was constructed based on the optimal

117
algorithm in order to generate the shortest pick-up sequence and route for the material
handling equipment. Two scenarios based on single item and multiple items are
explained for the measurement of the storage and retrieval system. Job scheduling
plays an important role in efficiency and productivity of warehousing operations. The
RFID-enabled warehousing system allows an item to be stored randomly at varying
locations wherever a storage place is available, i.e., the same type of items may have
multiple locations. Also, the RFID-enabled warehouse management system is capable
of identifying a dispatched item by issuing a priority to the selected item based on
assignment rules. A step by step process is used to explain all aspects of the
developed algorithm. Chapter 6 presents the development in the RFID-based
management system integration approach. The concept of integration enables RFID
components to be embedded within the mechanism infrastructure of the proposed
RFID-enable warehousing system in order to synchronise the warehouse operations
and update the inventory data in a real-time manner. This includes an integration of
hardware and software applications. The key components of the RFID inventory
management system for a pilot test were discussed and identified in this research
study. RFID system is used to allow users to improve accuracy, provides better
information/data control management, sufficient data processing speeds and
enhances security and reduce errors through automation. Chapter 7 illustrates the
experimental results through a pilot test based on the prototype of the proposed RFID-
based inventory management system. The developed RFID-based management
system can automatically detect and identify a number of locations of the ordered item
and display these in a row and column of a storage rack. Based on the algorithm
developed in the MATLAB environment, the system has the capability of issuing a
demand to dispatch the ordered item by assigning a priority to the selected item which
can also be displayed and manipulated via a GUI. The simulation result was generated
by the RFID-inventory management system according to the pre-defined selection
rules by issuing a priority to be given to the selected item. The mechanical control
system then initialises a demand to push the selected item onto an output conveyor
and this item will be transported along the RFID-guided route to the collection point.
The developed programming model of the RFID-enabled inventory management
system can generate the priority list of selected items to travel in a sequence from the
first item with a longest travel time to the last item with a least travel time, accordingly.
Furthermore, Chapter 8 illustrates the study aimed at developing a multi-objective
118
optimisation model for gaining a trade-off decision by measuring four conflicting
objectives: minimisation of the total cost, maximisation of capacity utilisation,
maximisation of service level to satisfy all demands of dispatching products from the
warehouse and minimisation of travel distance in the warehouse. By examining the
feasibility and applicability using a case study based on the develop model, the optimal
solutions can be subtracted from the objective functions. This also leads to an insight
into a compromised solution between conflicting objectives for decision makers.
Chapter 9 shows the cost analysis, benefits and limitations regarding RFID
implementation and a cost compression study between the previous barcode system
and the newly implemented RFID system based on a real case study. Chapter 10
provides a summary of this work and the conclusions and recommend the future work.

In supply chains and logistics sectors, accuracy of inventory data is essential as these
information data can be crucial for warehouse operations, SKU planning, management
and control of incoming and outgoing goods. To yield a desired performance, it is
necessary to ensure the operation of the warehouse work effectively. An effective real-
time inventory management system is also the foundation of a successful e-commerce
and online ordering that are growing at the exponential rate today. This PhD research
work was aimed to investigate and develop the design methodology and the
integration approach through which the RFID-enabled inventory management system
can perform an automatic availability check and update real-time information data of
ordered items from the warehouse database. Simultaneously, the RFID-embedded
mechanism of the AS/RR can automatically dispatch these items from the warehouse
without any human operation. Within the RFID-inventory management system, a
selection algorithm was developed to seek an optimal solution to determine a selected
item with a longest travel time and subsequently issue a priority to this item to be
dispatched first. This includes a transaction database which allows a manipulation of
RFID-tracked items under pre-defined rules by assigning a priority to one of selected
items.

The average travel time for the movement of the pusher was also determined by
considering both movement of the pusher with a constant velocity and with an
acceleration and declaration. Different analytical approaches showed that the
proposed analytical travel time models perform better in the constant velocity
compared to result of acceleration and declaration in which the largest deviations in

119
average travel time up to 4% were noticed with respect to acceleration and declaration.
The results, which were demonstrated in constant velocity and acceleration and
declaration, respectively, confirmed that there is not much significant deviation found
in the average travel time between constant velocity and acceleration and declaration.
Therefore, the simplified model for the travel time using constant velocity was
considered to be more realistic for this model.

A pilot test was carried out to examine the developed algorithm applied into the RFID-
based management system which, in theory, has a capability of interacting with the
mechanical control system of the RFID-embedded AS/RR. To synchronise these
systems to act coordinately, both the warehouse RFID-inventory management system
and the warehouse control system of the AS/RR need to be integrated through a
developed interface which allows an effective communication between these two
systems.

A multi-objective optimisation approach was also developed to optimise the design of

the proposed RFID-enabled automated warehousing system based on four conflicting
objectives in order to obtain trade-off solutions aimed at maximising the warehouse
capacity utilisation and service level in terms of satisfying all demands of dispatching
products from the warehouse, minimising the travel distance of products from a
storage rack to a collection point and the total cost required for implementing
RFID-related facilities into the proposed warehousing system. The study also includes
an approach to determine the optimal number of racks and collection points that
should be established when designing the proposed warehouse. At the first stage, an
integrated fuzzy solution approach was used to obtain Pareto-optima solutions. At the
second stage, TOPSIS was employed to reveal the best Pareto solution.
Subsequently, solution 6 in Table 16 was selected as the best solution since it
revealed the highest score. This score represents the nearest solution value from the
ideal solution and the furthest from the negative (worst) solution. Arguably, TOPSIS
does not involve or boost the optimisation since it can only help the decision makers
in selecting the final solution. The results showed that the developed multi-objective
model can be used as an aid for decision makers to obtain a cost-effective design of
an RFID-enabled automated warehousing system towards the optimisation of
conflicting objectives. A case study was also used for examining the feasibility and
applicability of the developed approach. The experiment shows that the developed

120
RFID-based management system can automatically detect and identify locations of
ordered items by displaying objects on GUI. Experimental results demonstrate the
applicability of the developed optimisation algorithm implemented into the RFID-based
inventory management system. It also shows that the RFID-inventory management
system can determine the optimal route for the pusher to travel in a sequence during
the ordering process that gives a minimal overall travel time for all the ordered items
to a specified collection point. Further, a study in cost analysis was conducted based
on the implementation of the RFID technology at the University of Portsmouth library.
This includes an outcome through an interview with the project manager for the
benefits and limitations when replacing the previous barcode system by the RFID
system.

121
10.1 The recommended future work

The pilot test was executed based on a limited number of RFID tagged items due to
the restriction of the research budget. In a real warehouse, however, there are a large
number of RFID-tagged items which need to be tracked and processed in a real time
manner through the integrated information systems. Within such a large scale
environment, RFID signal-overlapping and collision can cause a serious problem to
synchronise the operations of the entire automated warehouse. Thus, it is highly
recommended for the future work to employ a suitable distributed multi-channel reader
anti-collision approach for the proposed RFID-enabled warehouse system at a large
scale.

In recent years, it has been widely reported that the Internet of Things (loT) will be an
industrial revolution that may dramatically alter the ways in manufacturing, energy,
agriculture, transportation, supply chain and logistics and other industrial sectors. It
mayalso fundamentally transform how people will work through new interactions
between humans and machines. It may also change the basis of competition, redraw
industry boundaries and create a new wave of retail companies just as the current.
The IoT factory automation concept can also result in changes of supply chain
organization, operations and information management. One of the challenges is that
the warehouse management system (WMS) and the warehouse control system (WCS)
must be able to be not just integrated to establish a new autonomous system and but
also it supports new autonomous smart devices and multiple connectivity options.
RFID makes it possible to extend the fundamental abstraction of the world wide web
(www) to the physical world. Eventually, every interesting physical object can be
connected to a global network opening up the possibility for an Internet of Things (IoT),
although this expectation is still far from the existing world.

122
References

Agarwal, A., & Madan, A.K. (2015). A Review of Analytical Travel Time Models in
Automated Storage and Retrieval Systems. Journal of Basic and Applied Engineering
Research, 2, 1372-1376.

Ahmadi, A. Moghimi, H. Nezhadc, A.E. Agelidis, V.G. Sharaf, A.M. (2015). Multi-
objective economic emission dispatch considering combined heat and power by
normal boundary intersection method, Electric Power Systems Research, 129, 32–43.

Anderson, J.C. and Schroeder, R.G. (1984). Getting results from your MRP system.
Business Horizons, 27, 57-64

Andrew, L. M. (2006). A lean route to manufacturing survival. Assembly Automation,

26, 265– 272.

Ashayeri, J., & Gelders, L. F. (1985). Warehouse design optimisation. European

Journal of the Operational Research, 21, 285-294.

Ashayeri, J., Gelders, L. F. (1985). A microcomputer-based optimisation model for the

design of automated warehouses, International Journal of Production Research, 23,
4, 825–839.

Baars, H., Kemper, H.G., Lasi, H. & Siegel, M. (2008). Combining RFID technology
and business intelligence for supply chain optimisation scenarios for retail logistics.
Proceedings of the 41st Hawaii International Conference on System Sciences,(pp.1-
10)

Bafna, K. M., Reed, R. (1972). An Analytical Approach to Design of High-Rise tacker

Crane Warehouse Systems. Journal of Industrial Engineering, 4, 10, 8–14.

Bagchi, U., Guirida, A., O’Neill, L., Zeng, A., & Hayya, J. (2007). The effect of RFID on
inventory management and control. Springer series in advanced manufacturing. 71-
92.

Ballé, F, Ballé, M. (2005). Feel the force of flexible manpower. IET Manufacturing
Engineer, 84, 20-5.

Baysan, S. Ustundag, A. (2013). The Value of RFID: Benefits vs. Costs. Retrieved
from web address: https://books.google.co.uk/books?isbn=1447143450.

123
Bottani, E., & Rizzi, A. (2008). Economical assessment of the impact of RFID
technology and EPC system on the fast-moving consumer goods supply chain.
International Journal of Production Economics, 112, 548-569.

Brown, D. (2007). RFID Implementation. New York: McGraw-Hill Communications.

Brusey, J. and McFarlane, D.C. (2009). Effective RFID-based object tracking for
manufacturing. International Journal of Computer Integrated Manufacturing, 22, 638-
647.

Çakıcı, Ö.,E. Groenevelt, H. Seidmann, A. (2009). A case inquiry: can rfid help reduce
costs in medical imaging? Radiology Business Journal, 2, 50–52.

Cao, Y., Li, W. Song, W. and Chaovalitwongse, W., A. (2013). Collaborative Material
and Production Tracking in Toy Manufacturing. In IEEE 17th International Conference
on Computer Supported Cooperative Work in Design (CSCWD), 645–650.

Chao, C.C., Yang, J.M., & Jen, W.Y. (2007). Determining technology trends and
forecasts of RFID by a historical review and bibliometric analysis from 1991 to 2005.
Technovation, 27, 268-279.

Chawla, V., & Ha, D.S. (2007). An overview of passive RFID. IEEE Communication
Magazinee, 45, 11-17.

Chen, R.S., Tu, M.A., & Jwo, J.S. (2010). An RFID-based enterprise application
integration framework for real-time management of dynamic manufacturing
processes. International Journal of Advanced Manufacturing Technology, 50, 1217–
1234.

Chiandussi, G. Codegone, M. Ferrero, S. Varesio, F.E. (2012). Comparison of multi-

objective optimization methodologies for engineering applications, Computers &
Mathematics with Applications 63(5), 912-942.

Chomka, S. (2003). Radio tagging too pricey, claim many companies, Food
Manufacture, 78,13.

Chow, H.K.H., Choy, K. L., Lee, W. B., & Lau, K. C. (2006). Design of a RFID case-
based resource management system for warehouse operations. Expert Systems with
Applications, 30(4), 561-576.

124
Cusumano, M.A. (1994). The Limits of Lean. Sloan Management Review, 35 (4), 27-
32.

De Koster, R., Le-Duc, T., and Roodbergen, K.J. (2007). Design and control of
warehouse order picking: a literature review. European Journal of Operational
Research 182(2), 481-501.

Deavours, D.D. (2009). A Circularly Polarized Planar Antenna Modified for Passive
UHF RFID. IEEE International Conference on RFID.

Deb, K. (2001). Multi-Objective Optimisation using Evolutionary Algorithms, New York:

John Wiley & Sons.

Delaunay, C., Sahin, E., & Dallery, Y. (2007). A literature review on investigations
dealing with inventory management with data inaccuracies. RFID Eurasia. 1st Annual,
1-7.

Dennis, P. (2007). Lean production simplified. New York: Productivity Press.

Doerr, K.H., Gates, W.R., & Mutty, J. E. (2006). A hybrid approach to the valuation of
RFID /MEMS technology applied to ordnance inventor. International Journal of
Production Economics, 103, 726-741.

Dolgui, A., & Proth, J.M. (2008). RFID technology in supply chain management: State
of the art and perspectives. Proceedings of the 17th World Congress: The International
Federation of Automatic Control (pp. 4465-4475). Seoul, Korea.

Ebrahimi, S. Sharifian, S. Bastani, P (2015). Readiness of hospitals affiliated with

Tabriz University of Medical Sciences to Face Disasters. Journal of Health
Management, 3(1).

Ergen, E., Akinci, B., & Sacks, R. (2007). Life-cycle data management of engineered
to-order components using radio frequency identification. Advanced Engineering
Informatics, 21, 356-366.

Evangelatos, O., Samarasinghe, K. and Rolim, J. (2012). Evaluating design

approaches for smart building systems,”in Mobile Adhoc and Sensor Systems
(MASS), IEEE 9th International Conference , 1–7.

125
Ferguson, R. (2007). Berkeley's Mesh Network: Dust in the RFID Wind. eWeek.
Retrieved from: http://www.eweek.com

Gambon, J. (2009). Benefits Up and Down the Ski Slope. RfidJournal. Retrieved from:
http://www.rfidjournal.com

Gen, M. and Cheng, R. (1997). Genetic Algorithms and Engineering Design, New
York: John Wiley & Sons

Ghobbar, A. A. and Friend C. H. (2004). The material requirements planning system

for aircraft maintenance and inventory control: a note. Journal of Air Transport
Management 10, 217–221.

Grant, D.B., Lambert, D.M., Stock, J.R., and Ellram, L.M. (2006). Fundamentals of
logistics management. European edition. UK: McGraw-Hill Education

Gupta, S. and Jain, S. K. (2013). A literature review of lean manufacturing.

International Journal of Management Science and Engineering Management, 8, (4),
241–249.

Hamid, A. A., Agus, A. and Hassan, M., E. (1991). Computerised materials

requirement planning in manufacturing companies in Malaysia. International Journal
of Production Economics, 25, 73-79.

Harrop, P., & Das, R. (2010). Active RFID and sensor networks 2009-2019.

Hedgepeth, W.,O. (2006). RFID Metrics: Decision Making Tools for Today's Supply
Chains CRC Press, Boca Raton, FL.

Henrici, D., (2008). RFID Security and Privacy. Berlin. Springer-velag Berlin
Heidelberg.

Hines, P. Holwe, M. and Rich, N. (2004). Learning to evolve: A review of contemporary

lean thinking. International Journal of Operations & Production Management, 24
(9/10), 994-1011.

Hou, J.L., & Huang, C.H. (2006). Quantitative performance evaluation of RFID
applications in the supply chain of the printing industry. Industrial Management 8
Data Systems, 106, 96-123.

126
Hu, Y.H., Huang, S.Y., Chen, C., Hsu, W., Toh, A.C. Loh, C.K., & Tiancheng, S.
(2005). Travel time analysis of a new automatedstorage and retrieval system.
Computers & Operations Research, 32, 1515–1544.

Huang, S., Wang, Q., Batta, R., & Nagi, R. (2015). An integrated model for site
selection and space determination of warehouses. Computers & Operations
Research, 62, 169–176.

Huber, N., & Michael, K. (2007). Vendor perceptions of how RFID can minimise
product shrinkage in the retail supply chain. RFID Eurasia, 1st Annual (pp. 1-6).

Hunt, V. D., Puglia, M., & Puglia, A. (2007). A guide to radio frequency identification.
Technology Research Corporation.

Hwang, C. L. and Yoon, K. (1981). Multi Attribute Decision Making Methods and
Applications,”Springer-Verlag, Berlin Heidelberg, 259.

Hwang, H. Lee, SB. (1990). Travel time models considering the operating
characteristics of the storage and retrieval machine. Int J Prod Res, 28, 1779–1789.

Impinj. (2015) How Do RFID Systems Work? Impinj. Retrieved (2015). from:
http://www.impinj.com/resources/about-rfid/ how-do-rfid-systems-work

Impinj. (2015) RFID Standards. Impinj. Retrieved (2015). from:

http://www.impinj.com/resources/about-rfid/rfid-standards

Impinj. (2015). The different types of RFID systems. Impinj. Retrieved (2015) from:
http://www.impinj.com/resources/about-rfid/the-different-types-of-rfid-systems.

Jehng, W.K., Peng, S., & Huang, W.H.(2008). using LabVIEW to integrate RFID
system and database for supply chain efficiency improvement. International Journal
of Intelligent Control and Systems, 13, 189-195.

Jian, M.-S. (2010). RFID System Integration and Application Examples. Radio
Frequency Identification Fundamentals and Applications, Bringing Research to
Practice (chapter 14).

Jiménez López, M., Rodríguez Uría, M.V., Parra, M.A., & Bilbao, A. (2000). Solving a
possibilistic linear program through compromise programming. Mathware and Soft
Computing, 7, 175-184.

127
Kärkkäinen, M. (2003). Increasing efficiency in the supply chain for short shelf life
goods using RFID tagging. International Journal of Retail & Distribution Management,
31, 529 – 536.

Kim, E.Y., Ko, E., Kim, H., & Koh, C.E. (2008). Comparison of benefits of radio
frequency identification: Implications for business strategic performance in the US and
Korean retailers. Industrial Marketing Management, 37,797-806.

Kim, S., & Garrison, G. (2010). Understanding users' behaviors regarding supply chain
technology: determinants impacting the adoption and implementation of RFID
technology in South Korea. International Journal of Information Management, 30(5),
388-39.

Kodaka, K., Niwa, H., & Sugano S. (2009). Rotation-based dynamic localization at an
initial dead-zone avoidance stage on an rfid tag lattice. Journal of Robotics and
Mechatronics, 21(4), 460–468.

Landt, J. (2001). Shrouds of Time: The History of RPID. AIM Publications, Pittsburgh,
PA.

Landvater, D., (1985). How do you compare to 1,000 other companies? Infosystems,
32 (1).

Lee, S. G. Souza, R. de. Ong, E. K. (1996). Simulation modelling of a narrow aisle

automated storage and retrieval system (AS/RS) serviced by rail-guided vehicles.
Computers in Industry, 30,241-253.

Lee, W.-S., & Yu, J.-W. (2015). Selectable sectoral antenna array using a quadruple
feeding network for item-level tagging in UHF RFID applications. Microwave and
Optical Technology Letters, 57, 1523–1526.

Lee, Y.M., Cheng, F., & Leung, Y. T. (2004). Exploring the impact of RFID on supply
chain dynamics. In 2004 Winter Simulation Conference R .G. Ingalls, M. D. Rossetii,
J. S. Smith, and B. A. Peters (Eds.), 1145-1152.

Lefebvre, L.A., Lefebvre, E., Bendavid, Y., Wamba, S.F., & Boeck, H. (2006). RFID as
an enabler of B-to-B e-commerce and its impact on business processes: A pilot study

128
of a supply chain in the retail industry. In Proceedings of the 39th Hawaii International
Conference on System Sciences, 6, 1-10.

Lerher, T., Ekren, B.Y., Dukic, G., & Rosi, B. (2015). Travel time model for shuttle-
based storage and retrieval systems. International Journal of Advanced Manufacturing
Technology, 78, 9-12.

Lerher, T., Potrc, I., & Sraml, M. (2013). Designing automated warehouses by
minimising investment cost using genetic algorithms. Progress in material handling
research: 2010. Charlotte: The Material Handling Industry of America, 237–253.

Lerher, T., Potrc, I., Sraml, M., & Tollazzi, T. (2010). Travel time models for automated
warehouses with aisle transferring storage and retrieval machine. European Journal
of Operational Reserch,205, 571-583.

Lerher, T., Šraml, I., Borovinšek, M., & Potrč, I. (2013). Multi-objective optimisation of
automated storage and retrieval systems. Annals of faculty of mechanical engineering-
International journal of Engineering, 187-194.

Lerher,T., Sraml, M., Kramberger, J., Potrc, I., Borovinsek, M., & Zmazek, B.
(2006). Analytical Travel time model for multi aisle automated storage and retrieval
systems. The International Journal of Advanced Manufacturing Technology. 30, 340-
356.

Leung, Y.T., Cheng, F., Lee, Y.M., & Hennessy, J.J. (2007). A Tool Set for Exploring
the Value of RFID in a Supply Chain. Springer series in advanced manufacturing.

Lingle, R. (2007). RFID smart shelves and cabinets. Packaging World. Retrieved from:
http://www.healthcarepackaging.com/trends-and-issues/traceability-and-
authentication/rfid-smart-shelves-and-cabinets

Liu, C.M., & Chen, L.S. (2009). Applications of RFID Technology for improving
production efficiency in an integrated-circuit packaging house. International Journal of
Production Research, 47, 2203-2216.

Liu, C-M., Chen, L.S. (2009). Applications of RFID technology for improving production
efficiency in an integrated-circuit packaging house. International journal of production
research, 47, 2203-2216.

129
Liu, D., Mou, M., Lu, M., & Wu, Y. (2014). Travel Time Model of a New Compact
Storage System. International Journal of Control and Automation, 7, 447-460.

Liu, G., Mao, L., Chen, L., & Xie, S. (2014). Locatable-Body Temperature Monitoring
Based on Semi-Active UHF RFID Tags. Sensors, 14, 5952-5966.

Liu, G., Yu, W., and Liu, Y. (2006). Resource Management with RFID Technology in
Automatic Warehouse System. IEEE/RSJ International Conference On Intilegent
Robots And Systems, Beijing, 3706-3711.

Liu, Y., & Wang, D. (2009). An RFID middleware business process integration
framework based on EPC modeling and complex event processing. International
Conference on Computer Sciences and Convergence Information Technology, Seoul,
64-69.

Lwiki, T. Ojera, P.B. Mugenda, N.G. and Wachira, V.K. (2013). The Impact of Inventory
Management Practices on Financial Performance of Sugar Manufacturing Firms in
Kenya. International Journal of Business, Humanities and Technology, 3 (5), 75-85.

Lwiki, T., Ojera, P., B. Mugenda, N. G., Wachira, V., K. (2013). The Impact of inventory
Management Practices on Financial Performance of Sugar Manufacturing Firms in
Kenya. International Journal of Business, Humanities and Technology, 3, 75-85.

Ly, M. (2006). Automated Checkout System Using Radio Frequency ID Tags.

Monmouth University, New Jersey, United States. Readiness Assessment Report for
A&P Supermarkets.

Ma, H., Su, S., Simon, D., & Fei, M. (2015). Ensemble multi-objective biogeography-
based optimisation with application to automated warehouse scheduling. Engineering
Applications of Artificial Intelligence, 44, 79–90.

Ma, X., and Liu, T. (2011). The Application of Wi-Fi RTLS in Automatic Warehouse
Management System. In IEEE International Conference on Automation and Logistics
(ICAL), 64–69.

Marphy, Jr. P.R., & Wood, D.F. (2001). Contemporary Logistics. New Jersey: Pearson
Education.

130
McCathie, L. (2004). The advantages and disadvantages of barcodes and radio
frequency identification in supply chain management. University of Wollongong,
Australia.

McFarlane, D., Sarma, S., Chirn, J., Wong, C., & Ashton, K. (2003). Auto ID systems
and intelligent manufacturing control. Engineering Applications of Artificial Intelligence,
16, 365-376.

McGuinness, K. (2008). The Readiness of High-Technology Manufacturing

companies

Messac, A. Optimisation in Practice with MATLAB®: For Engineering Students and

Professionals, 1st Edition, 2015, Cambridge University Press.

Ming, K.L., Witold, B., and Stephen, C.H. (2013). RFID in the warehouse: A literature
analysis (1995-2010) of its applications, benefits, challenges and future trends.
International Journal of Production Economics. 145, 409-430.

Moeller, K. (2011). Increasing warehouse order picking performance by sequence

optimisation, Procedia - Social and Behavioral Sciences, 20, 177–185,.

Mohammed, A. Wang, Q. (2017). The fuzzy multi-objective distribution planner for a

green meat supply chain, International Journal Production Economics, 184, 47-58.

Mourtzis, D., Papakostas, N., Makris, S., Xanthakis, V., & Chryssolouris, G. (2008).
Supply chain modelling and control for producing highly customized products. CIRP
Annals - Manufacturing Technology, 57, 451-454.

Ne´meth,P.,Toth,L.,Hartvanyi,T. (2006). Adapting RFID in supply chains . IEEE 3rd

International Conference on Mechatronics ,263–266.

Nikitin, P., & Rao, K. V. S. (2008). Antennas and Propagation in UHF RFID Systems.
IEEE International Conference on RFID, 277–288.

Niu, N., Xu, L. D. and Bi, Z. (2013). Enterprise Information Systems Architecture-
Analysis and Evaluation. IEEE Transactions on Industrial Informatics 9, 2147–2154.
Osyczka, A. (1985) 7 – Multicriteria optimization for engineering design, Design
Optimization, Academic Press, 193–227.

131
Park, C.-S. (2014). Security Mechanism Based on Hospital Authentication Server for
Secure Application of Implantable Medical Devices. BioMed Research International,
2014, 543051.

Park, Y., & Yoo, J. (2008). An intelligent Middleware Platform and Framework for RFID
Reverse Logistics. International Journal of Future Generation Communication and
Networking.

Perry, R. F., Hoover, S. F., Freeman, D. R. (1983). Design of Automated Storage and
Retrieval Systems using Simulation Modeling, Institute of Industrial Engineers,
Atlanta, Georgia, ICAW Proceedings, 57–63.

Pigni, F., & Ravarini, A. (2008). RFID in the fashion industry: Evidences and lessons
learnt from an anti-counterfeiting project. In Proceedings of the Fourteenth Americas
Conference on Information Systems, (pp. 1-8).

Pishvaee, M.S. Razmi, J. (2012). Environmental supply chain network design using
multi-objective fuzzy mathematical programming, Applied Mathematical Modelling, 36
(8), 3433–3446.

Poon, K. T., Choy, K. L., & Lau, H. C. (2008). A RFID-based location tracking scheme
for inbound operations in warehouse environment. In: Management of Engineering &
Technology, 2008. PICMET (2008). Portland International Conference on IEEE, 872-
877.

Poon, T.C., Choy, K.L., Chow, H.K.H., Lau, H.C.W., Chan, F.T.S., & Ho, K.C. (2009).
A RFID case-based logistics resource management system for managing order-
picking operations in warehouses. Expert Systems with Applications, 36, 8277-8301.

Qiu, X., Huang, G.Q. (2013). Supply Hub in Industrial Park (SHIP): The value of freight
consolidation. Computers & Industrial Enginnring, 65, 16-27.

Rakotondrabe, M., Yassine, H. Clevy, C. and Lutz, P. (2011). Automation of Assembly

and Packaging at the Micro/Nano- Scale. In IEEE Conference on Automation Science
and Engineering (CASE), 1–5.

Ramesh, S., Kannan, S., & Baskar, S. (2013 ). Application of modified NSGA-II
algorithmto multi-objective reactive power planning. Appl. Soft Computing,12, 2013,
741–753.

132
RFID tags. (2015). Connectivity of the Internet of Things, Retrieved from:
https://www.google.co.uk/search?q=rfid+tags&biw=1920&bih=974&source=lnms&tb
m=isch&sa=X&sqi=2&ved=0ahUKEwiFzMi24N_PAhXjNJoKHQ-iDtwQ_AUIBygC.

Roberti, M. (2003). Analysis: RFID Wal-Marts network effect. (Online journal)

Retrieved from: http://www.cioinsight.com

Roberti, M. (2003). RFID’s Momentous Week, RFID journal, available at:

https://www.rfidjournal.com/articles/pdf?464

Roberti, M. (2004). Wal-Mart Begins RFID Rollout. RfidJournal. Retrieved from:

http://www.rfidjournal.com.

Rosenblatt, M. J., Roll, J. (1984). Warehouse design with storage policy

considerations, International Journal of Production Research, 22, 5, 809–821.

Ross, A.D., Twede, D., Clarke, R.H., & Ryan, R.H. (2009). A framework for developing
implementation strategies for a radio frequency identification (RFID) system in a
distribution centre environment. Journal of Business Logistics, 30, 157-183.

Sahin, E., & Dallery, Y. (2007). A literature review on the impact of inventory data
record inaccuracies on inventory management and the potential of the RFID
technology to tackle this issue. Ecole Centrale Paris, 2007.

Saygin, C. (2007). Adaptive inventory management using RFID data. The International
Journal of Advanced Manufacturing Technology, 32, 1045-1051.

Schenone, M., Mangano, G., Gabrina, S. (2016). Analysis of AS/RS travel times in
class-based storage environment through a simulation approach. Computational
Science and Systems Engineering, 170-177.

Seth, D., & Gupta, V. (2005). Application of value stream mapping for lean operations
and cycle time reduction: An Indian case study. Production Planning and Control, 16,
44–59.

Shaw, T. (2002). Reliability Measures for Highway Systems and Segments.

TRANSYSTEMS Corporation.

133
Singh, B., Garg, S. K., Sharma, S. K., & Grewal, C. (2010c). Lean implementation and
its benefits to production industry. International Journal of Lean Six Sigma, 1, 157–
168.

Song, H., Shin, V. and Jeon, M. (2012). Mobile Node Localization Using Fusion
Prediction-Based Interacting Multiple Model in Cricket Sensor Network. IEEE
Transactions on Industrial Electronics, 59, 4349–4359.

Sounderpandian, J., Boppana, R.V., Chalasani, S., & Madni, A.M. (2007). Models for
cost-benefit analysis of RFID implementations in retail stores. IEEE Systems Journal,
1, 105-114.

Squire, J. A. M. S., E. Baber, C. Absher, B. and Kendzia, T. (2014). Robo-Tic:

Development of a Tick-Eliminating Robot. IEEE Potentials, 33, 12–16.

Taguchi, M., & Mizuno, H. (2006). Analysis of Dead Zone of RFID System,
Proceedings of 2006 IEEE Antennas and Propagation Society International
Symposium, Albuquerque, 558, 4759-4762.

Ting, J.S.L., & Tsang, A.H.C. (2012). Design of an RFID-based Inventory Control and
Management System: A Case Study. The West Indian Journal of Engineering, 34, 70-
79.

To Adopt and Use Radio Frequency Identification (RFID): A Case Study of the Medical
Device Sector in Ireland. National University of Ireland Galway, Ireland. Retrieved
from: http://hdl.handle.net/10379/358.

Tompkins, J.A., White, J.A., Bozer, Y.A., Frazelle, E.H. and Tanchoco, J.M.A. (2003).
Facilities Planning, (NJ: John Wiley & Sons).

Turcu, C., Cerlincă, M., Cerlincă, T., Prodan, R., Turcu, T., Gîză, F., & Popa1, V.
(2007). An Integrated RFID-Based B2B System for Supply Chain Logistics and
Warehousing. International Journal of Computers and Communications, 1, 90-98.

Tzeng, S.F., Chen, W.H., & Pai, F.Y. (2008) Evaluating the business value of RFID:
Evidence from five case studies. International Journal of Production Economics, 112,
601-613.

134
Vasili, M.R. Tang, S.H. S Homayouni, M. and Ismail, N. (2006). Compersion of
different dwell point polices for split-platform automated storage and retrieval system.
International Journal of Engineering and Technology, 3, 91-106.

Vermesan, O., Friess. (2013). Internet of Things- Converging Technologies for Smart
Environments and integrated Ecosystems. Denmark, River publishers.

Wamba, S.F., Lefebvre, L.A., & Lefebvre, E. (2007). Integrating RFID technology and
the EPC network into a b2b retail supply chain: a step toward intelligent business
processes. Journal of Technology Management and Innovation, 2, 114-124.

Wang, H., Chen, S., & Xie, Y. (2010). An RFID-based digital warehouse management
system in the tobacco industry: a case study. International Journal of Production
Research, 48, 2513-2548.

Wang, L. Xu, L.D. Bi, Z. Xu, Y. (2014). Data cleaning for RFID and WSN integration,
IEEE Transactions on Industrial Informatics, vol.10, pp.408 – 418,.

Wang, L.C., Lin, Y.C., & Lin, P.H. (2007). Dynamic mobile RFID-based supply chain
control and management system in construction. Advanced Engineering Informatics,
21, 377-390.

Wang, Q., Alyahya, S., Bennett, N. & Dhakal, H. (2015). An RFID-enabled automated
warehousing system. International Journal of Materials, Mechanics and
Manufacturing, 3, 287-239.

Wang, Q., Mclntosh, R., & Brain, M. (2010). A new-generation automated warehousing
capability. International Journal of Computer Integrated Manufacturing, 6, 565-537.

Wang, X., & Adams, M. (2011). Warehousing and distribution centres. In Intermodal
transportation: Moving freight in a global economy, 1-22.

Werb, J., and Lanzl, C. (1998). Designing a Positioning System for Finding Things and
People Indoors. IEEE Spectrum, 35, 71–78.

Williams, S. Taylor, M. Irland. J. Metha, A. (2014). RFID Technology in Forensic

Evidence Management: An Assessment of Barriers, Benefits, and Costs. Retrieved
from web address: http://dx.doi.org/10.6028/NIST.IR.8030

135
Won, S., Golnaraghi, F. and Melek, W. W. (2009). “A Fastening Tool Tracking System
Using an IMU and a Position Sensor with Kalman Filters and a Fuzzy Expert System.”
IEEE Transactions on Industrial Electronics 56 (5): 1782–1792.

Xiaoyong, S., Chu, C.-C., Prabhu, B.S., & Gadh, R. (2007). On the Identification
Device Management and Data Capture via WinRFID Edge-Server. IEEE Systems
Journal – ISJ-8.

Xu , Z., Ming, X.G., Zhou, J. Song , W., He, L., & Miao Li. (2013). Management
optimisation based on dynamic SKU for RFID-enabled warehouse management in the
steel supply chain. International Journal ofProduction Research, 51, 2981-2996.

Yahong, H., Ying, H.S., Chuanyu, C., Wen-Jing, h., Cheong, T.A., & Kit, L.C. (2003).
Travel time analysis of a new Automated Storage and Retrieval System. IEEE
Emerging Technologies and Factory Automation,1, 75-81.

Yang, H. (2010). Integrated ZigBee RFID sensor networks for resource tracking and
monitoring in logistics management. Thesis. Loughborough University, UK.

Zhai, C. Zou, Z. Zhou, Q. Mao, J. Chen, Q. Tenhunena, H. Lirong, Z. and Xu, L. (2016).
A 2.4-GHz ISM RF and UWB Hybrid RFID Real-Time Locating System for Industrial
Enterprise Internet of Things (IoT), Enterprise Information System, published online.

Zhang, T., Wang, X., Liu, X., Chu, J., & Cui, P. (2010). Automotive green supply chain
management based on the RFID technology. (ICAMS 2010 - Proc. 2010). IEEE
International Conference on Advanced Management Science, 2, 617-619.

Zhou, W. (2008). RFID and item-level information visibility. European Journal of

Operational Research, 198, 252-258.

Zou, Z., Chen, Q. Uysal, I. and Zheng. L., R. (2014). Radio Frequency Identification
Enabled Wireless Sensing for Intelligent Food Logistics. Philosophical Transactions A
of the Royal Society, 372.

Zuehlke, D. (2010). Smartfactory-Towards a Factory-of-Things. Annual Reviews in

Control, 34, 129–138.

136
137
Appendix
Appendix A
Cumulative
Respondents Frequency Percent Valid Percent Percent
Valid Respondent_1 1 3.0 3.0 3.0
Respondent_2 1 3.0 3.0 6.1
Respondent_3 1 3.0 3.0 9.1
Respondent_4 1 3.0 3.0 12.1
Respondent_5 1 3.0 3.0 15.2
Respondent_6 1 3.0 3.0 18.2
Respondent_7 1 3.0 3.0 21.2
Respondent_8 1 3.0 3.0 24.2
Respondent_9 1 3.0 3.0 27.3
Respondent_10 1 3.0 3.0 30.3
Respondent_11 1 3.0 3.0 33.3
Respondent_12 1 3.0 3.0 36.4
Respondent_13 1 3.0 3.0 39.4
Respondent_14 1 3.0 3.0 42.4
Respondent_15 1 3.0 3.0 45.5
Respondent_16 1 3.0 3.0 48.5
Respondent_17 1 3.0 3.0 51.5
Respondent_18 1 3.0 3.0 54.5
Respondent_19 1 3.0 3.0 57.6
Respondent_20 1 3.0 3.0 60.6
Respondent_21 1 3.0 3.0 63.6
Respondent_22 1 3.0 3.0 66.7
Respondent_23 1 3.0 3.0 69.7
Respondent_24 1 3.0 3.0 72.7
Respondent_25 1 3.0 3.0 75.8
Respondent_26 1 3.0 3.0 78.8
Respondent_27 1 3.0 3.0 81.8
Respondent_28 1 3.0 3.0 84.8
Respondent_29 1 3.0 3.0 87.9
Respondent_30 1 3.0 3.0 90.9
Respondent_31 1 3.0 3.0 93.9
Respondent_32 1 3.0 3.0 97.0
Respondent_33 1 3.0 3.0 100.0
Total 33 100.0 100.0

A-1
1- Do you use database? If so, what types of database do you use?
Q1

Cumulative
Frequency Percent Valid Percent Percent

Valid Flat file 6 18.2 18.18 18.2

Structural database 21 63.6 60.6 81.8

N/A 4 12.1 12.12 93.9

Other 2 6.1 9.09 100.0

Total 33 100.0 100.0

2- What database do you use?

Q2

Cumulative
Frequency Percent Valid Percent Percent

Valid MySQL 6 18.2 17.65 18.2

Microsoft SQL 10 30.3 29.41 48.5

Oracle Database 7 21.2 20.59 69.7

Flat file 3 9.1 8.82 78.8

Microsoft Access 5 15.2 14.71 93.9

Other 2 6.1 8.82 100.0

Total 33 100.0 100.0

A-2
3- What do you currently store in your database? (Select all that apply)
Q3

Cumulative
Frequency Percent Valid Percent Percent

Valid Customer details 4 12.1 27.1 12.1

Products details 4 12.1 26.9 24.2

Sales 4 12.1 18.48 36.4

Suppliers details 5 15.2 21.74 51.5

Other 2 6.1 6.52 57.6

100.0

Total 33 100.0 100.0

A-3
4- Does your database link with enterprise resource planning (ERP)?
Q4

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 17 51.5 50.00 51.5

No 16 48.5 50.00 100.0

Total 33 100.0 100.0

5- Does your ERP have integrated warehouse management? If so, does it give you
enough information to measure operators’ productivity?
Q5

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 13 39.4 39.39 39.4

No 20 60.6 60.61 100.0

Total 33 100.0 100.0

A-4
6- Do you use material requirement planning (MRP) or other predictive planning?
Q6

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 15 45.5 44.12 45.5

No 18 54.5 55.88 100.0

Total 33 100.0 100.0

A-5
7- Is your database cloud based or on premise solution?
Q7

Cumulative
Frequency Percent Valid Percent Percent

Valid Cloud 20 60.6 60.6 60.6

On-premise 10 30.3 30.3 90.9

Other 3 9.1 9.09 100.0

Total 33 100.0 100.0

8- Do you use customer relationship management (CRM)?

Q8

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 16 48.5 47.06 48.5

No 17 51.5 52.94 100.0

Total 33 100.0 100.0

A-6
9- Do you use application programming interface (API)?
Q9

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 17 51.5 53.21 51.5

No 16 48.5 46.88 100.0

Total 33 100.0 100.0

10- Do you or the company you work for have a warehouse?

Q10

Cumulative
Frequency Percent Valid Percent Percent
Valid Yes 27 81.8 81.8 81.8

No 6 18.2 18.2 100.0

Total 33 100.0 100.0

A-7
11- What aspect of your operators’ performance do you need/want to measure?
Q11

Cumulative
Frequency Percent Valid Percent Percent

Valid Operator Accuracy 10 30.3 30.00 30.3

Operator Speed 10 30.3 30.00 60.6

Operator Efficiency 10 30.3 30.00 90.9

Other 3 9.1 10.00 100.0

Total 33 100.0 100.0

A-8
12- Do you use key performance indicator (KPI) to measure performance within the
warehouse?
Q12

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 16 48.5 47.06 48.5

No 17 51.5 52.94 100.0

Total 33 100.0 100.0

13- Have you heard of or came across an automated warehouse system?

Q13

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 20 60.6 60.61 60.6

No 13 39.4 39.39 100.0

Total 33 100.0 100.0

A-9
14- Do you consider your warehouse to be: manual, semi-automated or fully automated?
Q14

Cumulative
Frequency Percent Valid Percent Percent

Valid Manual 15 45.5 45.45 45.5

Semi-automated 11 33.3 33.33 78.8

Fully automated 7 21.2 21.21 100.0

Total 33 100.0 100.0

A-10
15- Does your business require real time inventory?
Q15

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 21 63.6 63.64 63.6

No 12 36.4 36.36 100.0

Total 33 100.0 100.0

16- Do you use barcode, voice picking or any other technology within your warehouse?
Q16

Cumulative
Frequency Percent Valid Percent Percent

Valid Barcodes 17 51.5 50.00 51.5

Voice picking 14 42.4 41.18 93.9

Please Specify 2 6.1 8.82 100.0

Total 33 100.0 100.0

A-11
17- Have you heard of radio frequency identification (RFID)?
Q17

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 23 69.7 70.59 69.7

No 10 30.3 29.41 100.0

Total 33 100.0 100.0

A-12
18- Have you heard of RFID based warehouse management system?
Q18

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 15 45.5 44.12 45.5

No 18 54.5 55.88 100.0

Total 33 100.0 100.0

19- How do you rate the following attributes in your warehouse management system?
Q19

Cumulative
Frequency Percent Valid Percent Percent

Valid Accuracy 5 15.2 73.12 15.2

Efficiency 5 15.2 64.67 30.3

Operator speed 4 12.1 62.67 42.4

Stock loss 4 12.1 69.9 54.5

All 15 45.5 100.0

Total 33 100.0 100.0

A-13
20- What is the estimated percentage in human errors to pick up 100 items in a manual
warehouse?
Q20

Cumulative
Frequency Percent Valid Percent Percent

Valid 1%-25% 17 51.5 51.5 51.5

26%-50% 5 15.2 15.2 66.7

51%-75% 4 12.1 12.1 78.8

76%-100% 7 21.2 21.2 100.0

Total 33 100.0 100.0

21- Does your business provide ecommerce service?

Q21

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 15 45.5 44.12 45.5

No 18 54.5 55.88 100.0

Total 33 100.0 100.0

A-14
22- Do you incentivise the use of ecommerce by means of cost reduction?
Q22

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 16 48.5 47.06 48.5

No 17 51.5 52.94 100.0

Total 33 100.0 100.0

A-15
23- What delivery options do you provide? (select all that apply)
Q23

Cumulative
Frequency Percent Valid Percent Percent

Valid Same day delivery 9 27.3 31.48 27.3

Next day delivery 15 45.5 40.74 72.7

Three days’ delivery 8 24.2 22.22 97.0

Other 1 3.0 5.56 100.0

Total 33 100.0 100.0

24- Do your customers require flexibility for delivery window?

Q24

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 21 63.6 61.76 63.6

No 12 36.4 38.24 100.0

Total 33 100.0 100.0

A-16
25- Is the speed of delivery an important factor?
Q25

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 30 90.9 88.24 90.9

No 3 9.1 11.76 100.0

Total 33 100.0 100.0

A-17
26- Do you provide reserve and collect service?
Q26

Cumulative
Frequency Percent Valid Percent Percent

Valid Yes 20 60.6 58.82 60.6

No 13 39.4 41.18 100.0

Total 33 100.0 100.0

27- How do you rate future generation automated warehouse system?

Q27

Cumulative
Frequency Percent Valid Percent Percent
Valid 30% 1 3.0 3.0 3.0

31%-50% 2 6.1 6.1 9.1

51%-75% 3 9.1 9.1 18.2

76%-100% 27 81.8 81.8 100.0

Total 33 100.0 100.0

A-18
Appendix B
Table 20.B: Summary of related studies.

Authors Year Model Findings / Developments

Wang et al 2010 RFID Defined a set of event-condition-action

(ECA) rules to improve the feasibility and
flexibility of the digital warehouse
management system in a tobacco company

Ming et al 2013 RFID Examined the previous and current status

and a future trend in terms of RFID-related
applications, benefits and obstacles
providing a generic view in the field for
researchers.

Chow et al 2006 RFID Proposed an RFID-based resource

management system (RFID-RMS), used for
logistics providers to enhance warehouse
operations by means of tracking and
optimising resources utilisation

Poon et al 2009 RFID Simplification of RFID implementation

procedure, an improvement in visibility of
warehouse operations and an
enhancement of productivity of the
warehouse

Ting et al 2012 RFID Proposed an RFID-based inventory control

and management system for manufacturing
enterprises. It aimed to collect accurate
real-time data relating to transactions of
physical stock items for enhancing product
life cycle management

Uckun et al 2008 RFID Proposed a model to maximise the

expected profit by seeking an optimum
number of warehouses using the RFID
techniques to minimise the total costs

Anssens et al 2010 RFID Proposed a solution to improve the

reliability of RFID reader in warehouse, to
synchronize reader’s beams.

Sahin et al 2007 RFID Examined a solution for the issue of data

inaccuracy of warehouse inventories

B-1
Cakici et al 2011 RFID Analysed the incremental benefits of using
RFID tags over barcodes for
pharmaceutical inventory management

Xu et al 2013 RFID Introduced a useful and comprehensive

warehouse management optimization
method based on dynamic storage keeping
unit (SKU) management

Liu et al 2009 RFID Implemented an electronic control system

to improve the production efficiency in a
local integrated-circuit (IC) packaging
house.

Turcu et al 2007 RFID Examined the potential benefits generated

by integration of an RFID system with
business-to-business (B2B) applications
and subsequently proposed an integrated
RFID-B2B system.

Jehng et al 2008 RFID Investigated an automatic conveyor system

on which the material flow can be monitored
and traced in a real time manner using
RFID tags

Yoo et al 2008 RFID The work proposed a network infrastructure

that supports logistic activities for the entire
process from a recycling of product
materials to a disposal by adding reverse
logistics to the existing forward logistics.

Rouwenhorst et al. 2000 Stated that efficiency and effectiveness in

any distribution network is largely
determined by the operation of the nodes
in the network, i.e., the warehouses

Xue et al 2011 RFID Introduced the robot technology into the

automated warehouse, which aims to
improve S/R efficiency

Rashad et al 2011 Wireless Improved existing warehouse management

communication system by implementing wireless
technology for controlling ASAR system
during the operation.
Brusey et al 2009 RFID The research presented an approach for
making best use of RFID data evaluated in
the perspective of a laboratory

B-2
 manufacturing system that produces
customized gift boxes.

Xiaoguang et al 2008 RFID&WSN Implemented an adaptive communication

framework of the warehouse management
based on RFID and Wireless Sensor
Network (WSN)

Li et al 2011 Introduced a robot automatic system

intelligent warehouse based on RFID
technology

Chen et al 2010 RFID Proposed an RFID-based enterprise

application integration (EAI) approach for
the real-time management of dynamic
manufacturing processes. A prototype
system was developed to demonstrate the
applicability of the proposed method in a
shop floor environment

Irani et al 2010 RFID Provided a comprehensive and systematic

survey of the literature pertaining to RFID
related research issues in order to ascertain
the current ‘state of play’ of the field along a
number of dimensions. The objectives
are:1) to analyse the distribution/trend of
RFID research across subject category,
source titles (journals), geographical
locations, document types and year of
publications; 2) to determine the frequently
published authors and productive
institutions for conducting RFID related
research; 3) to explore the trend of
topics/research issues and utilised
methods; and finally (4) to synthesis the
existing research to develop a research and
may guide the practitioner for implementing
and managing RFID applications in both
public and private sectors

Srujana et al 2013 RFID Developed an RFID-based library

management system using MATLAB

Ahsan et al 2010 RFID Aimed at developing a model for mobile

application in order to track patients’
movements in a hospital. Also the main

B-3
 components of an RFID system are
identified

Brezovnik et al 2015 AS/RS Conducted a study to provide a solution for

AS/RS planning and operation. It
simultaneously address structural,
operational and control features of the
AS/RS

Ye Lu 2008 RFID Proposed a new RFID general framework

based on middle ware data cleaning in bus
park information integration system.

Chen et al 2010 RFID Proposed an RFID-based enterprise

application integration (EAI) approach for
the real-time management of dynamic
manufacturing processes. A prototype
system was developed to demonstrate the
applicability of the proposed method in a
shop floor environment

Herageu et al 2005 Warehouse Proposed a mathematical model and a

design heuristic algorithm that jointly determine
product allocation to the functional areas in
the warehouse as well as the size of each
area using data readily available to a
warehouse manager

Kabadurmus et al 2007 RFID Addressed an effective methodology to

reveal appropriate operational settings for
any RFID system. The purpose of this study
is (1) to present an organized, concise
methodology for identification of influential
factors of any RFID system (2) to evaluate
operational parameters on RFID controlled
conveyor system.

Hu et al 2005 Travel time Presented a new kind of S/R mechanism

that enables AS/RS to efficiently handle
very heavy loads. Also developed a
continuous travel-time model for the new
AS/RS under the stay dwell point policy.

Liu et al 2014 UHF RFID Demonstrated a combined way to perform

tags both RFID real-time remote monitoring of
body temperature and location of the
respective body.

B-4
Al-Naima et al 2012 RFID Proposed a simulation system for the
vehicle traffic congestion estimation
(VTCE) based on RFID. The RFID reader
will read the vehicle tags and transfer the
necessary information to a database in a
central computer system (CCS).

De Koster et al 2008 AS/RS Considered a newly designed compact

three-dimensional automated storage and
retrieval system (AS/RS). The system
consists of an automated crane taking care
of the pallets movements in the horizontal
and vertical direction

Lerher et al 2010 Travel time Presented analytical travel time model for
the computation of travel time for
automated warehouses with the aisle
transferring S/R machine (in continuation
multi-aisle AS/RS). These models consider
the operating characteristics of the storage
and retrieval machine such as acceleration
and deceleration and the maximum velocity

Lerher et al 2015 Travel time Presented analytical travel time model for
the computation of travel (cycle) time for
shuttle-based storage and retrieval
systems (in continuation SBS/RS). The
proposed model considers the operating
characteristics of the elevators lifting table
and the shuttle carrier, such as acceleration
and deceleration and the maximum velocity

Yang et al 2015 AS/RS Examined the joint optimization of storage

location assignment and storage/retrieval
scheduling in multi-shuttle automated
storage/retrieval systems (AS/RSs) under
shared storage, in which the reuse of empty
location yielded by retrieval operation is
allowed.

Kuan et al 2005 RFID Describes the experience of developing a

RFID system coupling with an inventory
management system, with currently is
installed on the RAPIDTECH company in
Taiwan.

B-5
Ballestín et al 2013 RFID The objective of this reserch is threefold, to
propose and compare different offline and
online policies for the scheduling of
warehouse operations, to design a tool that
allows the decision maker to compare
policies and environments without putting
them into practice, and to study the benefits
that can be obtained if RFID is used in a
particular type of warehouse. Authors
developed a stylised model that captures
and generalises the main characteristics of
the structure, routing and sequencing
operations of a given real warehouse.

Badole et al 2013 RFID Presented the impact of RFID in warehouse

management of perishable products. It
provides mathematical framework to
assess the benefits of RFID in warehouse
management. It helps management in the
variety of ways including improvement in
receiving and shipping processes,
reduction in cycle counting efforts,
reduction in stock outs/excess inventory,
decreased counterfeiting, decreased
returns, and reduction in inventory loss due
to shrinkage and obsolescence.

B-6
Appendix C
C.1 MATLAB
MATLAB is a language in an interactive environment for developing algorithms,
analysing and visualizing data and performing numeric computation. Users can import
data into MATLAB from files, other applications or external devices. Once the data is
in MATLAB, it can be ‘explore’ and ‘analyse’ it through built in engineering and
mathematical functions and plots and visualisations. The MATLAB language support
the vector and matrix operations that a fundamental to scientific and engineering
problems. Command can be executed one at the time providing immediate results;
this lets a user to explore multiple approaches in obdurate to an optimal solution. A
user can create scripts and functions to re-use and automate such work.
Developments tools allow users to implement algorithms efficiently and optimise their
performance. MATLAB provides the features of a traditional programming language
as well as layout tools for designing costume, graphical user interfaces. Add-on tool
boxes extend the MATLAB environment to solve problems in a range of applications
including signal processing in communications, video and image processing, control
design and computational biology. The use of MATLAB can significantly increase
efficiency in development, by keeping the code simple, readable and short. MATLAB
facilitates data analysis and display of information (Daniel et al., 2007). Additional
features include useful toolboxes, including database and instrument control settings.
The MATLAB layout is illustrated in the following Figure 1.C.1.1

Figure 48.C.1.1 Screen shot of the MATLAB layout.

C-1
The desktop includes these panels:

1. Current Folder: Access your files.

2. Command Window: Enter commands at the command line, indicated by the
prompt (>>).
3. Workspace: Explore data that you create or import from files.
4. Command History: View or rerun commands that you entered at the command
line.
Table 21.C.1.1: MATLAB programming codes.

% Part 1: Database
% The database contains two sections, for which section 1 is starts with
"Tag"
% stands for the barcode of the items and section 2 is starts with "Name"
that
% indentify the what exactly the items are

% Section 1: Barcode
Tag_1_1=char('ABCDMMMM20140101', 'DEFGSTUR00000000', 'GHIYMMMM20140201',
'RTTUCCCC20140302', 'HELLYYYY20140304', 'HEWWCHCH20140404',
'HSALMMMM20140510', 'HALYGGGG20140709', 'SUNSHINE20140606');
Tag_1_2=char('ABCDWXYZ20140909', 'DEFGSSSS20140505', 'GHSTKLMN00000000',
'RTHANAYZ20140507', 'HILAHOLD20140919', 'HAKIMMMM20140616',
'MAMAJJJJ20141011', 'ALYAHYAS00000000', 'SUNRISEE20141214');
Tag_1_3=char('ABGFCCCC20141111', 'DSTGYYYY20141212', 'GHSTCHCH20140620',
'ABCDSTUR00000000', 'HILABBBB20141215','BANDFFFF20140617',
'BABAMAMA20141012','ALYAWWWW20140123', 'WXYZMMMM20140101');
Tag_1_4=char('BCDCSSSS20150101', 'DRMRJJJJ20141230', 'GHYGZZZZ20140808',
'RTTIGGGG20140709', 'WORLDWIL00000000','BADEOOOO20140620',
'TELLMEYA20141015', 'MALYAHYA20140131', 'RSTUFZXY20140111');
Tag_1_5=char('ABDUZZZZ20140506', 'DDEFRRTT20141225', 'GHIYAHYA00000000',
'RAMIRAMI20140208', 'HANWCHCH20140219', 'DREAMMMM20141221',
'DINNGGGG20140810', 'RAAFHAHA20140710','MAJETTTT20130909');
Tag_1_6=char('ABGFSSSS20141113', 'DSTGJJJJ20131213', 'KLHYZZZZ20140622',
'ABCAGGGG20140525', 'WARABBBB20141216','BANDIIII20140619',
'RGBAYYYY20141031','ALBAIIII20140405', 'HANIGMMA20140415');
Tag_1_7=char('ABGGFFFF20140617', 'SALEYYYY20141202', 'SAMICCCC20131010',
'ERUPECAR20131012', 'HIBYTTTT20141204','BANDYLTD20140719',
'DADYWWWW20131011','ABRAHIMS00000000', 'HABICHCH20140520');
Tag_1_8=char('ABGGSSSS20140617', 'ALBAJJJJ20141213', 'MADIZZZZ20131016',
'MACCAMAD20131028', 'RIYADHGF20141218','BURICHCH20140722',
'DAMAOOOO20131014','AHMEIIII20131101', 'WASEBBBB20141215');
Tag_1_9=char('OSAMABIN20140911', 'TAIFYYYY20131214', 'ALYAHYAS00000000',
'TBOOCCCC20140526', 'HEATYYYY20141116','PORTCHCH20140621',
'SOUATTTT20141013','BRIGHTON20130405', 'BATHBRIS20130417');

% Section 2: Name
% Bearing in mind the length of each char has been modified by manually
% adding some space at the end. This is to make sure each variable has
% exactly same length and it is important and necessary for the data
% processing. The reason is that different items have different names and

C-2
% for what the chars amount of each item is different. Mismatch length will
% result in programme crashing down.
% Further improvement will be considered in the future
Name_1_1=char('Milk ', 'pen ', 'Milk ', 'Coke ','Yogurt
','Cheese ','Milk ','Eggs ','Ice-cream');
Name_1_2=char('Sugar ','Salt ','CD ','Juice ','Honey
','Milk ','Jam ','Cable ','Fanta ');
Name_1_3=char('Cock ','Yogurt ','Cheese ', 'DVD ','Bread
','Beans ','Orange ','Water ','Milk ');
Name_1_4=char('Salt ', 'Jam ', 'Butter ', 'Eggs
','Headphone','Olives ','Chicken ','Tomato ','Biscuits ');
Name_1_5=char('Butter ','Pasta ','Speaker ','Coffee ','Cheese
','Milk ','Eggs ','Banana ','Tea ');
Name_1_6=char('Salt ', 'Jam ', 'Butter ', 'Eggs ','Bread
','Oil ','Yogurt ','Oil ','Yogurt ');
Name_1_7=char('Beans ', 'Yogurt ', 'Cock ','Bread ','TEA
','Orange ','Water ','Charger ','Cheese ');
Name_1_8=char('Salt ', 'Jam ', 'Butter ', 'Eggs ','COFFEE
','Cheese ','Olives ','Oil ','Bread ');
Name_1_9=char('Milk ', 'Yogurt ', 'Game ', 'Coke ','Yogurt
','Cheese ','TEA ','Lamb ','Onion ');
PP=[9;1]; % pusher position

%
% Section3: Define the value of the corresponding characteristics
Hf=0.4; % height of shelf unit
rack=9; % rack number of a shelf
L=0.5; % length of shelf unit
Lsc=0.25; % length between the edge of the spiral and the conveyor system
Lss=0.5; % length of each single spiral
Lsp=1.5; % length between end of spiral and collection point
D=0.5; % depth of shelf unit
Vc=0.2; % speed of conveyor
Vp=0.3; % speed of pusher movement
Vpp=0.1; % speed of a pusher to push the item out
Vs=0.2; % speed of spiral movement

%--------------------------------------------------------------------------
% The time characteristics need to be calculated are as follows:
% Tm: the time for the pusher to move to the ordered item.
% Tp: time for the pusher to push out the ordered item.
% Ts: time for the item to move on the conveyor to the spiral.
% Ti: time of the movement of the item on the spiral.
% Te: time between the edge (end) of the spiral and the collection point.
% Tt: total time to get the ordered item.
%==========================================================================
% Part 2: Data Processing
disp('Please Input The Name of Ordered Item!')
OrderedItem='Milk '; % Change the name of ordered item, this syntax
will be improved later

%--------------------------------------------------------------------------
% Find and save the location information which contains the ordered item
No=0; % Starting number of ordered item
Location=[]; % For storing the location information of ordered item
for i=1:9
for j=1:9
eval(['Name=Name_1_' num2str(i) '(j,:);'])
if isequal (Name , OrderedItem)==1
No=No+1; % the location number increase, final value of No
give

C-3
 % the information of how may locations contains the
% ordered item
Lo=[i;j]; % the location which contains the ordered item
Location=[Location Lo]; % save the location information
else
No=No+0;
end
end
end
%--------------------------------------------------------------------------
% Find and save the information of expiry date for all location which
% contains the ordered item
Barcode=[]; % To save the barcode of ordered item at different locations
ExpiryDate=[]; % To save the expiry date of ordered item at different
% locations in string format
for k=1:No
eval(['Tag=Tag_1_' num2str(Location(1,k)) '(' num2str(Location(2,k))
',:);' ])
Barcode=[Barcode;Tag];
Date=Tag(1,end-7:end); % extract the information of expiry date
ExpiryDate=[ExpiryDate;Date];
end
%--------------------------------------------------------------------------
% Check if the ordered item has expiry date?
if isequal(ExpiryDate(1,:)>0, [1 1 1 1 1 1 1 1])==1
% If the ordered item has expiry date,
disp('The ordered item has expiry date!')
disp('Milk!')
Expiry=[]; % To save the expiry date of ordered item at different
locations
% in number format
if No==1
% if No=1, it means only one location contains the ordered item
Location1=Location;
%--------------------------------------------------------------------
% Calculate the computation time
Tm=sqrt((PP(1)*Hf-Location1(1)*Hf)^2+(PP(2)*L-Location1(2)*L)^2)/Vp;
Tp=D/Vpp;
Ts=((Location1(2)-0.5)*L+Lsc)/Vc;
Ti=(rack-Location1(1))*Lss/Vs;
Te=Lsp/Vc;
Tt=Tm+Tp+Ts+Ti+Te;
%--------------------------------------------------------------------
disp('The time to get the ordered item is:')
disp(Tt)
PP=Location1;
else
for m=1:No
A=str2num(ExpiryDate(m,:)); % transfer the string to number for
facilitaing
% the expiry date study
Expiry=[Expiry A];
end
NewData=sort(Expiry); % sort the array in ascending order
if NewData(1)==NewData(end)
disp('The ordered item at different locations have same expiry
date!')
Loca=[];
TT=[]; % empty array for saving the computation time at
multiple locations
for m=1:No

C-4
 if Expiry(m)==NewData(1)
Loc=Location(:,m);
Loca=[Loca Loc];
end
end
for n=1:length(Loca)
Location1=Loca(n);
%--------------------------------------------------------
% Calculate the computation time
Tm=sqrt((PP(1)*Hf-Location1(1)*Hf)^2+(PP(2)*L-
Location1(2)*L)^2)/Vp;
Tp=D/Vpp;
Ts=((Location1(2)-0.5)*L+Lsc)/Vc;
Ti=(rack-Location1(1))*Lss/Vs;
Te=Lsp/Vc;
Tt=Tm+Tp+Ts+Ti+Te;
%--------------------------------------------------------
TT=[TT Tt];
end
T_min=min(TT);
disp('The time to get the ordered item is:')
disp(T_min)
for n=1:length(Loca)
if TT(n)==T_min
Location2=Loca(:,n); % Samllest computation time
location
end
end
PP=Location2;
else
disp('The ordered item at different locations have different
expiry dates!')
if sum(Expiry==NewData(1))==1
% the NewData(1) is the minimun date so it is also the one
% which is more close to expire date. If the value of
% sum(ExpiryDate==NewData(1)) is 1, it means only one location
% contains the ordered item with cloest expire date. If not,
% there will be multiple locations.
for m=1:No
if Expiry(m)==NewData(1)
Location1=Location(:,m);
end
end
%------------------------------------------------------------
% Calculate the computation time
Tm=sqrt((PP(1)*Hf-Location1(1)*Hf)^2+(PP(2)*L-
Location1(2)*L)^2)/Vp;
Tp=D/Vpp;
Ts=((Location1(2)-0.5)*L+Lsc)/Vc;
Ti=(rack-Location1(1))*Lss/Vs;
Te=Lsp/Vc;
Tt=Tm+Tp+Ts+Ti+Te;
%------------------------------------------------------------
disp('The time to get the ordered item is:')
disp(Tt)
PP=Location1;
else
Loca=[];
TT=[]; % empty array for saving the computation time at
multiple locations
for m=1:No

C-5
 if Expiry(m)==NewData(1)
Loc=Location(:,m);
Loca=[Loca Loc];
end
end
for n=1:length(Loca)
Location1=Loca(:,n);
%--------------------------------------------------------
% Calculate the computation time
Tm=sqrt((PP(1)*Hf-Location1(1)*Hf)^2+(PP(2)*L-
Location1(2)*L)^2)/Vp;
Tp=D/Vpp;
Ts=((Location1(2)-0.5)*L+Lsc)/Vc;
Ti=(rack-Location1(1))*Lss/Vs;
Te=Lsp/Vc;
Tt=Tm+Tp+Ts+Ti+Te;
%--------------------------------------------------------
TT=[TT Tt];
end
T_min=min(TT);
disp('The time to get the ordered item is:')
disp(T_min)
for n=1:length(Loca)
if TT(n)==T_min
Location2=Loca(:,n); % Samllest computation time
location
end
end
PP=Location2;
end
end
end
else
% If the ordered item have no expiry date,
disp('The ordered item have no expiry date!')
if No==1
% if No=1, it means only one location contains the ordered item
Location1=Location;
%--------------------------------------------------------------------
% Calculate the computation time
Tm=sqrt((PP(1)*Hf-Location1(1)*Hf)^2+(PP(2)*L-Location1(2)*L)^2)/Vp;
Tp=D/Vpp;
Ts=((Location1(2)-0.5)*L+Lsc)/Vc;
Ti=(rack-Location1(1))*Lss/Vs;
Te=Lsp/Vc;
Tt=Tm+Tp+Ts+Ti+Te;
%--------------------------------------------------------------------
disp('The time to get the ordered item is:')
disp(Tt)
PP=Location1;
else
Loca=[];
TT=[]; % empty array for saving the computation time at multiple
locations
for m=1:No
if Expiry(m)==NewData(1)
Loc=Location(:,m);
Loca=[Loca Loc];
end
end
for n=1:length(Loca)

C-6
 Location1=Loca(:,n);
%---------------------------------------------------------------
-
% Calculate the computation time
Tm=sqrt((PP(1)*Hf-Location1(1)*Hf)^2+(PP(2)*L-
Location1(2)*L)^2)/Vp;
Tp=D/Vpp;
Ts=((Location1(2)-0.5)*L+Lsc)/Vc;
Ti=(rack-Location1(1))*Lss/Vs;
Te=Lsp/Vc;
Tt=Tm+Tp+Ts+Ti+Te;
%---------------------------------------------------------------
-
TT=[TT Tt];
end
T_min=min(TT);
disp('The time to get the ordered item is:')
disp(T_min)
for n=1:length(Loca)
if TT(n)==T_min
Location2=Loca(:,n); % Samllest computation time location
end
end
PP=Location2;
end
end

C-7
MATLAB toolbox supports communications using open database connectivity (ODBC)
with a compatible database such as Microsoft SQL. An ODBC is a middleware
application programming interface (API) that can be used for accessing the database
management system as shown in Figure 2.C.1.2

Figure 49.B.1.2 Screen shot of manipulating the RFID data using MATLAB functions.

C-8
Appendix D
D.1 GUI using MATLAB
MATLAB functions via the use of a graphical user interface development environment
(GUIDE) which offers a range of tools to make GUIs, including a figure window that
contains text graphics, panels, buttons, and menus, which enables a more simplified
process of development for UIs. The GUIDE Layout Editor can be used to populate a
UI simply through clicking and dragging the UI components, e.g. panels, buttons, text
fields, sliders, and other features within the layout area as required. Also menus or
context menus for the UI can be adjusted in this way, changing the components look
and feel, whether aligned parallel to other objects or listing them hierarchically. A
program file is automatically generated by GUIDE. In this research, there are two
methods of displaying processed data using the MATLAB environment; one way it can
be displayed is in a command window, and another way is using a GUI. The benefit of
displaying this information in the command window is for presenting the data for
correlating its accuracy, whereas the GUI has the advantage of highly effective visuals.
In this project, the first way was used for near field as according to the timeframe of
implementing this system and the GUI is used to display both near field and far field,
generically.

Table 22.D.1.1: Graphical user interface (GUI) programming code.

function varargout = GUI_Office(varargin)

% GUI_OFFICE M-file for GUI_Office.fig
% GUI_OFFICE, by itself, creates a new GUI_OFFICE or raises the
existing
% singleton*.
%
% H = GUI_OFFICE returns the handle to a new GUI_OFFICE or the handle
to
% the existing singleton*.
%
% GUI_OFFICE('CALLBACK',hObject,eventData,handles,...) calls the local
% function named CALLBACK in GUI_OFFICE.M with the given input
arguments.
%
% GUI_OFFICE('Property','Value',...) creates a new GUI_OFFICE or
raises the
% existing singleton*. Starting from the left, property value pairs
are
% applied to the GUI before GUI_Office_OpeningFcn gets called. An
% unrecognized property name or invalid value makes property
application
% stop. All inputs are passed to GUI_Office_OpeningFcn via varargin.

D-1
%
% *See GUI Options on GUIDE's Tools menu. Choose "GUI allows only one
% instance to run (singleton)".
%
% See also: GUIDE, GUIDATA, GUIHANDLES

% Edit the above text to modify the response to help GUI_Office

% Last Modified by GUIDE v2.5 10-Mar-2015 19:23:04

% Begin initialization code - DO NOT EDIT

gui_Singleton = 1;
gui_State = struct('gui_Name', mfilename, ...
'gui_Singleton', gui_Singleton, ...
'gui_OpeningFcn', @GUI_Office_OpeningFcn, ...
'gui_OutputFcn', @GUI_Office_OutputFcn, ...
'gui_LayoutFcn', [] , ...
'gui_Callback', []);
if nargin && ischar(varargin{1})
gui_State.gui_Callback = str2func(varargin{1});
end

if nargout
[varargout{1:nargout}] = gui_mainfcn(gui_State, varargin{:});
else
gui_mainfcn(gui_State, varargin{:});
end
% End initialization code - DO NOT EDIT

% --- Executes just before GUI_Office is made visible.

function GUI_Office_OpeningFcn(hObject, eventdata, handles, varargin)
% This function has no output args, see OutputFcn.
% hObject handle to figure
% eventdata reserved - to be defined in a future version of MATLAB
% handles structure with handles and user data (see GUIDATA)
% varargin command line arguments to GUI_Office (see VARARGIN)

% Create the data to plot

Data_warehouse=imread('warehouse.jpg','jpg');
image(Data_warehouse);
colormap bone
hold on

% Choose default command line output for GUI_Office

handles.output = hObject;

% Update handles structure

guidata(hObject, handles);

% UIWAIT makes GUI_Office wait for user response (see UIRESUME)

% uiwait(handles.figure1);

% --- Outputs from this function are returned to the command line.
function varargout = GUI_Office_OutputFcn(hObject, eventdata, handles)
% varargout cell array for returning output args (see VARARGOUT);
% hObject handle to figure
% eventdata reserved - to be defined in a future version of MATLAB

D-2
% handles structure with handles and user data (see GUIDATA)

% Get default command line output from handles structure

varargout{1} = handles.output;

function edit1_Callback(hObject, eventdata, handles)

% hObject handle to edit1 (see GCBO)
% eventdata reserved - to be defined in a future version of MATLAB
% handles structure with handles and user data (see GUIDATA)

% Hints: get(hObject,'String') returns contents of edit1 as text

% str2double(get(hObject,'String')) returns contents of edit1 as a
double

% --- Executes during object creation, after setting all properties.

function edit1_CreateFcn(hObject, eventdata, handles)
% hObject handle to edit1 (see GCBO)
% eventdata reserved - to be defined in a future version of MATLAB
% handles empty - handles not created until after all CreateFcns called

% Hint: edit controls usually have a white background on Windows.

% See ISPC and COMPUTER.
if ispc && isequal(get(hObject,'BackgroundColor'),
get(0,'defaultUicontrolBackgroundColor'))
set(hObject,'BackgroundColor','white');
end

% --- Executes on button press in search_pushbutton.

function search_pushbutton_Callback(hObject, eventdata, handles)
% hObject handle to search_pushbutton (see GCBO)
% eventdata reserved - to be defined in a future version of MATLAB
% handles structure with handles and user data (see GUIDATA)

prompt = {'Enter the key words of wanted item:'}; % Display the text
answer = inputdlg(prompt); % output the typed text
var=answer(1);
% save('Input','var')
load database
[m n]=size(databset_warehouse);
location=[];
for k=1:m
Da=databset_warehouse(k,4);
if isequal(var,Da)==1
location=[location k];
end
end
date=min(str2num(cell2mat(databset_warehouse(location,2))));
if date>0
A=str2num(cell2mat(databset_warehouse(:,2)))==date;
B=find(A==1);
% row=str2num(cell2mat(databset_warehouse(B,1)));
% col=str2num(cell2mat(databset_warehouse(B,6)));
row=cell2mat(databset_warehouse(B,1));
col=cell2mat(databset_warehouse(B,6));
else
date=0;

D-3
 B=location(1);
row=cell2mat(databset_warehouse(B,1));
col=cell2mat(databset_warehouse(B,6));
end
result=[location(1) row col B length(location) date];
save('result','result')
plot(col,row,'bo', 'MarkerSize', 10); % plot and mark the wanted location
in the map

% --- Executes on button press in display_pushbutton.

function display_pushbutton_Callback(hObject, eventdata, handles)
% hObject handle to display_pushbutton (see GCBO)
% eventdata reserved - to be defined in a future version of MATLAB
% handles structure with handles and user data (see GUIDATA)
unilogo=imread('LogoUni.jpg');
load database
load result
name=cell2mat(databset_warehouse(result(1),4));
ava=num2str(result(5));
loca=num2str([result(2) result(3)]);
if result(6)>0
EDate=num2str(result(6));
ID=num2str(cell2mat(databset_warehouse(result(4),3)));
else
EDate='No expiry date for this item.';
ID=num2str(cell2mat(databset_warehouse(result(4),3)));
end

h=msgbox({['Item name: ', name]; ['Item ID: ',ID]; ['Expiry date: ' EDate];
...
['Item availability: ', ava]; ['Item location: ', loca]}, 'Item
Information','custom',unilogo);

% --- Executes on button press in clear_pushbutton.

function clear_pushbutton_Callback(hObject, eventdata, handles)
% hObject handle to clear_pushbutton (see GCBO)
% eventdata reserved - to be defined in a future version of MATLAB
% handles structure with handles and user data (see GUIDATA)
delete(findobj(gca,'Type','line','Marker','+'));
delete(findobj(gca,'Type','line','Marker','o'));

% --- Executes during object creation, after setting all properties.

function axes1_CreateFcn(hObject, eventdata, handles)
% hObject handle to axes1 (see GCBO)
% eventdata reserved - to be defined in a future version of MATLAB
% handles empty - handles not created until after all CreateFcns called

% Hint: place code in OpeningFcn to populate axes1

% --- Executes on button press in pushbutton4.

function pushbutton4_Callback(hObject, eventdata, handles)
% hObject handle to pushbutton4 (see GCBO)
% eventdata reserved - to be defined in a future version of MATLAB
% handles structure with handles and user data (see GUIDATA)

% Display all the tagged item in the map

plot([50,100,140,130,190],[40,90,50,135,100],'r+', 'MarkerSize', 10);

D-4
Appendix E
E.1 Hardware component
RFID-based warehousing management system consists of the following elements:
RFID Reader
RFID tags
Serial to USB converter
Personal computer.
Arduino is an open-source physical computing platform based on a simple I/O board
and a development environment that implements the Processing/Wiring language.
Arduino can be used to develop stand-alone interactive objects or can be connected
to software on your computer (e.g. Flash, Processing, MaxMSP). The Arduino has the
ability to be installed to any type of operating system and could be powered and
programmed via the built in USB connection. The programming functionality using C++
enhances the ability to use this board. The Arduino is an open source platform, which
means that it has large number of technical documentation and support.

Figure 50.E.1.1: Arduino Uno Rev.3 (ATmega328).

Near field communication (NFC) is a set of standards for smartphones and similar
devices to establish radio communication with each other by touching them together
or bringing them into close proximity, usually no more than a few centimetres. Present
and anticipated applications include contactless transactions, data exchange, and
simplified setup of more complex communications such as Wi-Fi.

E-1
 Figure 51.E.1.2: RFID 13.56 MHz / NFC Card. (Tags)

Overview

ID12-LA and the ID20-LA series are small footprint 2.8-5.0volt reader modules that
support ASCII, Wiegand26 and Magnetic ABA Track2 data formats. The modules are
pin and function compatible with the ID2/12/20 series.

Figure 52.E.1.3: The ID-12 reader module.

RFID reader (ID-12) with Arduino
The RFID ID-12 is a simple reader that reads 125 kHz (EM 4001 or compatible) tags.
The datasheet provides data structure, wiring and pinout details as shown in Figure
6.E.1.4.

E-2
 Figure 53.E.1.4: The Arduino interface.
Table 23.E.1.1: Basic description of the testing equipment.

Reader specification

Name Impinj RFID reader

Frequency 902~928 MHz

Antenna far-field antenna

Antenna Passive

Frequency 902~928 MHz

Antenna Linear

E-3
 Tag

Nature Passive (middle)

RFID tags used in this study

The security requirements in an RFID system depends on the type and sensitivity of
data some RFID tags may have additional security layers and require passwords or
identification codes. This may affect functionality of RFID tag. Most applications of the
RFID tag employ a unique identifier which can be located via its row tag database
(RTD). As the majority of system users have access to a network, the RFID information
relating to the tagged items can also be accessed diverse types and amounts of RFID
data are restricted by limitations of the database operations and settings. RFID tag
selection is part of any RFID implementation. There are several HF RFID tags
standards are used, which are as follow:
1. RFID passive tag (near-field communication)

ISO 15693 standard for tracking items.

ECMA-340 and ISO/IEC 18092 standard for Near Field Communications (NFC).
ISO/IEC 14443 A and ISO/IEC 14443 standards for MIFARE technology, which is
used in smart cards.
JIS X 6319-4 standard for FeliCa which is used in a smart card system for electronic
money cards.
2. RFID passive tag (far-field communication)

E-4
Table 24.E: Characteristics of passive RFID tags used in this research for far-field
antenna.

Device type Class 1 Generation 2 passive UHF RFID transponder

Air interface protocol EPCGlobal Class1 Gen2 ISO 18000-6C
Operational frequency Global 860-960MHz
IC type Impinj Monza 4QT & Impinj Monza 4E.
Memory configuration With Monza 4QT:EPC 128 bit; User 512 bit; TID 96 bit
With Monza 4E:EPC 496 bit; User 128 bit; TID 96 bit
Read Range (EU on plastic up to 12,5 m/ 41 ft, EU on cardboard
up to 11 m/ 36 ft, US on plastic up to 12 m/ 39 ft and
US on plastic up to 10,5 m/ 34 ft).
Application surface Non-metallic surfaces.
materials

E-5
Appendix F
F.1 Visual Studio Project

Table 7.F.1.1 shows the SDK programming code (C#) for processing RFID data in
the MVS environment.

Table 25.F.1.1: RFID application using MVS.

using RfidTracker.Domain;

namespace RfidTracker.FluentApi
{
public class RfidAntennaBuilder
{
private static RfidAntenna _rfidAntenna;

public RfidAntennaBuilder()
{
_rfidAntenna = new RfidAntenna();
}

public static RfidAntennaBuilder DefaultLocation()

{
return new RfidAntennaBuilder();
}

public static RfidAntenna Build()

{
return _rfidAntenna;
}

public RfidAntennaBuilder WithName(string name)

{
_rfidAntenna.Name = name;
return this;
}

public RfidAntennaBuilder WithLocation(Location location)

{
_rfidAntenna.Location = location;
return this;
}

public RfidAntennaBuilder WithLocation(string floor, string room, string

shelf)
{
var location = new Location();
location.Floor = floor;
location.Room = room;
location.Shelf = shelf;
_rfidAntenna.Location = location;
return this;
}
}
}

F-1
using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
using RfidTracker.DataLayer;
using RfidTracker.Communications;

namespace RfidTracker.Reader
{
public static class RfidControllerFactory
{
public static IRfidController GetRfidController(IUnitOfWork unitOfWork)
{
if (unitOfWork.UseRfidSimulation)
{
return new RfidControllerSimulator(unitOfWork);
}
return new RfidController(unitOfWork.SystemProperties.ToList());
}
}
}

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Threading.Tasks;
using RfidTracker.DataLayer;
using RfidTracker.Domain;
using RfidTracker.Communications;

namespace RfidTracker.Reader
{
public class Program
{
private static IUnitOfWork _unitOfWork;
private static IRfidController _rfidController;

static void Main(string[] args)

{
DisplayWelcomeMessage();
Initialise();
RfidController.RunReader();

Console.WriteLine("");
Console.WriteLine("Finished: Press any key to stop the program.");
Console.WriteLine("");
}

private static IUnitOfWork UnitOfWork

{
get { return _unitOfWork; }
}

private static IRfidController RfidController

{
get { return _rfidController; }

F-2
 }

private static void DisplayWelcomeMessage()

{
Console.ForegroundColor = ConsoleColor.DarkGreen;
Console.WriteLine("**************************");
Console.WriteLine("Welcome to the RFID Reader");
Console.WriteLine("**************************");
Console.WriteLine(string.Empty);
Console.WriteLine("Press <enter> to start.");
Console.ReadLine();
Console.WriteLine(string.Empty);
Console.WriteLine("Starting . . .");
Console.WriteLine(string.Empty);
}

private static void Initialise()

{
_unitOfWork = new UnitOfWork();
_rfidController = RfidControllerFactory.GetRfidController(UnitOfWork);
_rfidController.RfidTagRead += RfidControllerRfidTagRead;
_rfidController.RunReader();
Console.ReadLine();
}

private static void RfidControllerRfidTagRead(object source,

RfidTagReadEventArgs rfidTagReadEventArgs)
{
var message = string.Format("{0} {1} {2} {3}",
rfidTagReadEventArgs.RfidTag, rfidTagReadEventArgs.Hostname,
rfidTagReadEventArgs.IpAddress, rfidTagReadEventArgs.PortNumber);
Console.WriteLine(message);

UnitOfWork.ProcessRfidReadEvent(rfidTagReadEventArgs);

}
}
}

<?xml version="1.0" encoding="utf-8"?>

<configuration>
<configSections>
<!-- For more information on Entity Framework configuration, visit
http://go.microsoft.com/fwlink/?LinkID=237468 -->
<section name="entityFramework"
type="System.Data.Entity.Internal.ConfigFile.EntityFrameworkSection, EntityFramework,
Version=6.0.0.0, Culture=neutral, PublicKeyToken=b77a5c561934e089"
requirePermission="false" />
</configSections>

<connectionStrings>
<!-- To change the database, change the Catalog to RfidTrackerWarehouse -->
<add name="RfidTrackerDatabase"
connectionString="Data Source=(LocalDb)\Projects;Initial
Catalog=RfidTrackerWarehouse;Integrated Security=SSPI"
providerName="System.Data.SqlClient" />
</connectionStrings>

<startup>

F-3
 <supportedRuntime version="v4.0" sku=".NETFramework,Version=v4.5" />
</startup>
<entityFramework>
<defaultConnectionFactory
type="System.Data.Entity.Infrastructure.LocalDbConnectionFactory, EntityFramework">
<parameters>
<parameter value="v11.0" />
</parameters>
</defaultConnectionFactory>
<providers>
<provider invariantName="System.Data.SqlClient"
type="System.Data.Entity.SqlServer.SqlProviderServices, EntityFramework.SqlServer" />
</providers>
</entityFramework>
</configuration>

// Set the latest location

Console.WriteLine("Updating Location for '{0}'", trackedItem.Lastname);
trackedItem.LatestLocationId = rfidAntenna.LocationId.Value;

// Add the tracked item event

var tracketItemEvent = new TrackedItemEvent();
tracketItemEvent.Id = Guid.NewGuid();
tracketItemEvent.EventDate = System.DateTime.Now;
tracketItemEvent.LocationId = rfidAntenna.LocationId.Value;
tracketItemEvent.TrackedItemId = trackedItem.Id;
TrackedItemEvents.Add(tracketItemEvent);

try
{
SaveChanges();
}
catch (DbEntityValidationException e)
{
foreach (var eve in e.EntityValidationErrors)
{
Console.WriteLine("Entity of type \"{0}\" in state \"{1}\" has the
following validation errors:",
eve.Entry.Entity.GetType().Name, eve.Entry.State);
foreach (var ve in eve.ValidationErrors)
{
Console.ForegroundColor = ConsoleColor.DarkRed;
Console.WriteLine("- Property: \"{0}\", Error: \"{1}\"",
ve.PropertyName, ve.ErrorMessage);
Console.ForegroundColor = ConsoleColor.DarkGreen;
}
}
throw;
}
catch (Exception ex)
{
Console.ForegroundColor = ConsoleColor.DarkRed;
Console.WriteLine(ex.Message);
Console.ForegroundColor = ConsoleColor.DarkGreen;
}
}

}
}

F-4
Workflow: Read Tags

Read Tags. Record the following information:

 Host name
 IP address
 Port number
 RFID EPC value

Get the RFID reader that contains the host name

or IP address that its read.
Reader not found
SELECT * From RfidReader
Where HostHame=@hostname
Or IP address =@ IpAddress
Report an error

Get the RFID antenna that is linked to the RFID

reader and has the port number its read. Antenna not found

SELECT * From RfidAntenna

Where Rfidreaderid=@rfidreaderid
Report an error
And PortNumber=@PortNumber

Get the tracked item with an EPC Value that its

read. Tracked item not
found
SELECT * From Trackeditem
Where EpcValue=@epcValue Report an error

We have all the information are needed.

Problem updating
INSERT RfidTagEvent (hostname, database
ipaddress,portnumber,epcvalue,date)
INSERT TrackedItemEvent (Trackeditemid Report an error
,locationid)
UPDATE Trackeditem and set LatestLocationid

Figure 54.F.1.1: Workflow acquires an RFID tag using a MVS environment.

F-5
Appendix G
G.1 Opening the RFID Database
RFID Database: The RFID database is implemented using Microsoft SQL Server:
RFID Tracker Warehouse. Open SQL Server Management Studio to view the
databases. The Server name is (localdb)\Projects.

Figure 55.G.1.1: Screen shot of the Microsoft SQL layout.

After connecting to the database server (using the screen above), created databases
will be shown as RFID Tracker Warehouse.

Figure 56.G.1.2: Screen shot of the RFID database tracker warehouse within SQL
server.

G-1
Ensure that the application is setup to seed the warehouse database.
Goto the RfidTracker.DataLayer project
Open the file ‘DatabaseInitialiser’
select to seed the Warehouse – just make sure only one of them is commented out /
set to run.

Figure 57.G.1.3: Screen shot of the RFID database Initialiser within MVS.

Compile the Application and Run it.

Compile by selecting Build  Build Solution from the main menu
Then this could be run by selecting the run ‘Google Chrome’ from the main toolbar if
applicable.
F.2 Running the RFID reader console
This section shows you how to run the RFID reader console application.
Step 01 – Set the Startup Project
Open Visual Studio and load the Project

Figure 58.G.2.1: Screen shot of the Microsoft Visual Studio layout.

In the Solution Explorer right click the RFID Tracker. Reader project and select ‘Set
as Start-up Project’

G-2
 Figure 59.G.2.2: screen shot of Setting up an RFID tracker reader as Start-up
Project.
Step 02 – Application Configuration.
The RfidTracker.Reader project has its own config file which tells it which database
to use.
Ensure it is correct

Figure 60.G.2.3: screen shot of Application configuration within MVS.

G-3
Step 03 – Build and Run
Build  Build Solution from the main menu.
The run using F5
The simulation engine simulates some tracked items moving.
When it completes you can verify the data in the database.

Figure 61.G.2.4: The MVS simulator window.

Table 26.G.1.1: Processing data and record in the RFID database.

USE [RfidTrackerWarehouse]
-- Get the last 10 records from the tag event table
SELECT TOP 10 * FROM [dbo].[RfidTagEvent] ORDER BY [DateAndTime] DESC
-- Get the last 10 records from the tracked item event table
SELECT TOP 10 * FROM [dbo].[TrackedItemEvent] ORDER BY [EventDate] DESC
-- Get the items that moved!
SELECT *
FROM [dbo].[TrackedItem]
WHERE [Id] IN (SELECT TOP 10 [TrackedItemId]
FROM [dbo].[TrackedItemEvent]
ORDER BY [EventDate] DESC)

G-4
Appendix H
H.1 RFID Database Schema
The following section describes the RFID database table for the future work.
System Property
This table stores the system settings
SELECT * FROM dbo.SystemProperty
Users and Roles
Stores users and roles, these are used for storing the passwords. Passwords are
encrypted. There is only one role at this stage – admin. The system can easily be
extended to have more roles.

RFID Readers, Antennae and Locations

A reader will be linked to many antennas; each antenna is linked to one location.

H-1
 Figure 62.H.1.1: Information stored in the RFID database.
Tracked Items
The system stores tracked items. (Inventory items) Each tracked item has a latest
location. When an RFID event occurs (i.e. when we read an RFID tag), we record the
tag event and then update the tracked item event table and the latest location. This
allows us to record the movements of tracked items and also record the latest location
of an item.

Figure 63.H.1.2: Tracked Items within RFID database.

Accessing the Database
Use SQL Server Management Studio to access and look at the data in the database.
SELECT – to read data
UPDATE – to update a record
INSERT – to insert records
DELETE – to delete records.

H-2
Appendix I
H.1 Simulation
Table 27.I.I.1: simulation programming code for single item

x=1:5;
y=[2 4 7 3 5]
z=[20 44 50 12 27];

plot3(x,y,z,'o',x:1,y:[2],z:[20],'*')

grid on

legend('items','Least travel time')

xlabel('Row of the ordered item')
ylabel('Column of the ordered item')
zlabel('Travel time (Min)')
title('Total travel time across the grid in a single-item')

Table 28.I.1.2: Simulation programming code for multiple items

x=1:10;
y=[2 4 7 3 5 2 3 6 4 8]
z=[20 44 50 12 27 25 33 45 55 37];

plot3(x,y,z,'o')

grid on

legend('Selected items')
xlabel('Row of the ordered items')
ylabel('Column of the ordered items')
zlabel('Travel time (Min)')
title('Total travel time across the grid')

Table 29.I.1.3: Example of analysis of items selection process.

x=[5 4 3]
y=[3 4 5]
z=[23 44 15];

plot3(x,y,z,'o',x,y,z,'*')
grid on
legend('Selected items','First item to arrive')

xlabel('Row of the ordered items')

ylabel('Column of the ordered items')
zlabel('Travel time(Min)')
title('Total travel time across the grid in multiple items')

I-1
x=[5 4 3]
y=[3 4 5]
z=[23 44 15];

plot3(x,y,z,'o',x:5,y:3,z:23,'*')
grid on
legend('Selected items','First item to arrive')

xlabel('Row of the ordered items')

ylabel('Column of the ordered items')
zlabel('Travel time(Min)')
title('Total travel time across the grid in multiple items')

Figure 17.I.1.1 shows the MATLAB simulation result which determines an item with a
least travel time among ordered items of the same type to a collection point in an
AR/RR. It indicates that the item at location (1, 2) has a least travel time of 20 seconds
and thus this item was selected and dispatched to a collection point. The result was
generated by the RFID-inventory management system according to the pre-defined
selection rules and the system subsequently issues a priority to be given to the
selected item. The mechanical control system then initialise a demand to push the
selected item onto an output conveyor and this item will be transported along the RFID-
guided route to the collection point.

I-2
Figure 64.I.1.1: Determination of a least travel time among ordered items in the
same type.

I-3