Available online at [Link].
com
Available online at [Link]
ScienceDirect
ScienceDirect
Procedia online
Available Computer
at Science 00 (2018) 000–000
[Link]
Procedia Computer Science 00 (2018) 000–000 [Link]/locate/procedia
[Link]/locate/procedia
ScienceDirect
Procedia Computer Science 151 (2019) 691–698
The 2nd International Conference on Emerging Data and Industry 4.0 (EDI40)
The 2nd International
AprilConference
29 – May 2,on2019,
Emerging Data
Leuven, and Industry 4.0 (EDI40)
Belgium
April 29 – May 2, 2019, Leuven, Belgium
Reliability Assessment in the Context of Industry 4.0:
Reliability Assessment in the Context of Industry 4.0:
Data as a Game Changer
Data as a Game Changer
Sanja Lazarova-Molnara* and Nader Mohamedb
aMærsk
Sanja Lazarova-Molnara* and Nader Mohamedb
Mc-Kinney Møller Institute, University of Southern Denmark, Odense, Denmark
aMærsk bMiddleware Technologies Laboratory, Pittsburgh, PA, USA
Mc-Kinney Møller Institute, University of Southern Denmark, Odense, Denmark
bMiddleware Technologies Laboratory, Pittsburgh, PA, USA
Abstract
Abstract
Reliability is the measure of the likelihood that a product, system or service will perform its intended function adequately for a
Reliabilityperiod
specified is the of
measure of thereliability
time. Low likelihood ofthat a product, system
a manufacturing or service
system, besideswill
the perform its intended
costly repairs function adequately
and replacements, for a
also implies
specifiedproduction,
reduced period of time. Low reliabilitysignificantly
and consequently, of a manufacturing
reduced system,
profits. besides
Therefore,the itcostly repairs
is very and replacements,
important to have a way alsotoimplies
assess
reduced production,
reliability, and consequently,
as a key performance metric forsignificantly
manufacturingreduced profits.
systems, andTherefore, it is systems
cyber-physical very important to have
in general. a waydeveloped
The newly to assess
reliability, asand
information a key performance metric
communication for manufacturing
technologies systems, andbecoming
that are increasingly cyber-physical
part ofsystems in general.
the current The newly
and future developed
manufacturing
information
systems, bothand communication
allow and invite for technologies that are
more sophisticated increasingly
approaches becoming
to assessing part ofofthe
reliability current and systems
manufacturing future manufacturing
as opposed to
systems,
the both allow
traditional expertand invite for more approaches.
knowledge-based sophisticatedInapproaches
this paper,towe assessing
describereliability of manufacturing
the significance systems
of evaluating as opposed
reliability to
for the
the traditional
progress expert knowledge-based
and acceptance of the Industry approaches. In thisaspaper,
4.0 technologies, well aswe thedescribe the significance
new directions of evaluating
and possibilities reliability
for enhanced for the
reliability
progress that
analysis and these
acceptance of the Industry
new technologies can4.0 technologies,
provide. Finally,aswewell as thean
provide new directions
overview and implications
of the possibilities of
forthese
enhanced
novelreliability
ways of
analysis that
analyzing these new
reliability technologies
in the can provide.
context of Industry 4.0. Finally, we provide an overview of the implications of these novel ways of
analyzing reliability in the context of Industry 4.0.
© 2019 The Authors. Published by Elsevier B.V.
© 2019 The Authors. Published by Elsevier B.V.
© 2019
This The
is an openAuthors.
accessPublished by Elsevier
article under the CC B.V.
BY-NC-ND license ([Link]
This is an open access article under the CC BY-NC-ND license ([Link]
This is an
Peer-review open access
under article
responsibilityunder
of the
the
Peer-review under responsibility of the Conference CC BY-NC-ND
Conference Program
Program license ([Link]
Chairs.
Chairs.
Peer-review under responsibility of the Conference Program Chairs.
Keywords: manufacturing systems; reliability; data collection; challenges; opportunities
Keywords: manufacturing systems; reliability; data collection; challenges; opportunities
* Corresponding author. Tel.: +45-40121112;.
* E-mail slmo@[Link]
address:author.
Corresponding Tel.: +45-40121112;.
E-mail address: slmo@[Link]
1877-0509 © 2019 The Authors. Published by Elsevier B.V.
This is an open
1877-0509 access
© 2019 Thearticle under
Authors. the CC BY-NC-ND
Published license ([Link]
by Elsevier B.V.
Peer-review under
This is an open responsibility
access of the Conference
article under CC BY-NC-NDProgram Chairs.
license ([Link]
Peer-review under responsibility of the Conference Program Chairs.
1877-0509 © 2019 The Authors. Published by Elsevier B.V.
This is an open access article under the CC BY-NC-ND license ([Link]
Peer-review under responsibility of the Conference Program Chairs.
10.1016/[Link].2019.04.092
692 Sanja Lazarova-Molnar et al. / Procedia Computer Science 151 (2019) 691–698
2 Author name / Procedia Computer Science 00 (2018) 000–000
1. Introduction
Reliability evaluates the likelihood that a product, system or service will perform its intended function adequately
for a specified period of time. Low reliability of a manufacturing system, besides the costly repairs and
replacements, also implies reduced production, and consequently, reduced profits. Reliability of manufacturing
systems has been of great importance for many decades [1, 2]. This fact has not changed throughout the years; on
the contrary, reliability has maintained increased its significance as a vital measure of the operability of
manufacturing facilities with their increasing complexities [3, 4]. What has changed, though, is that data that can
enable more efficient and accurate reliability analysis of systems have now become even more easily obtainable.
Therefore, a great portion of the expert knowledge that was typically leading and supporting reliability analysis has
gotten an opportunity to be both enhanced and validated.
Reliability of a manufacturing system affects a number of other efficiency performance metrics, such as
production costs, maintenance costs and spare parts management costs, ultimately affecting productivity and
revenue, as low reliability implies that machines have frequent breakdowns, and thus, frequent interruptions in
manufacturing processes. Related metrics, such as availability, also take into account systems’ and components’
downtimes due to repairs, so some of these performance metrics provide a more comprehensive way to assess. With
the emergence of Industry 4.0, there are new opportunities of automating processes that are relevant for performing
reliability analysis of manufacturing systems, allowing for new types of data to be collected on all relevant
phenomena, including humans, and human-machine interactions. The development of new technologies and data
infrastructures also enables the development of validation approaches for all historically used reliability models, and
their enhancements or calibrations.
Moreover, recently predictive and prescriptive maintenance of machines has been identified as the most
important application of industrial data analytics, as stated by Columbus in [5]. Since reliability is one of the more
relevant measures to evaluate maintenance schedules, strategies or sequences of actions, consequently accurate
reliability analysis and assessment are just as important. Throughout history we have been witnesses of a large
number of oversimplified and wrong reliability models that have been used for making costly decisions [6, 7].
Decisions made based on inaccurate models can often be damaging to the performance of a system, affecting also
both its safety and its security.
The goal of this paper is to discuss the significance of reliability analysis in the context of Industry 4.0 and
provide directions for enhancing the processes related to reliability analysis of manufacturing facilities, given the
new Information and Communication Technologies (ICT) developments.
We begin by providing background and preliminaries in Section 2, followed by a study of the meaning of
reliability in the Industry 4.0 context in Section 3. In Section 4, we discuss the implications of the new opportunities
for performing reliability analysis in the context of Industry 4.0, taking into consideration the latest technological
developments.
2. Background
In this section we elaborate the context of Industry 4.0 and its principles, as well as summarize the advancements
in reliability analysis for manufacturing systems.
2.1. On Industry 4.0
Industry 4.0 is the term that has been adopted to denote the latest digital industrial revolution [8]. The major drive
for this industrial revolution is the data, i.e. the convenience and ease with which it is being collected and analyzed.
This is, of course, facilitated by the large storage and computing power and the development of advanced and
efficient methods for analyzing data and transforming it into decisions. All industries have been affected by these
technologies and advancements, thus, utilizing them to enhance production by a large number of performance
criteria [9], such as profit, costs, customer experience, energy efficiency, customer lifetime values and customer
loyalty, to name a few [10-14]. The emergence and development of Industry 4.0 is currently creating significant
Sanja Lazarova-Molnar et al. / Procedia Computer Science 151 (2019) 691–698 693
Author name / Procedia Computer Science 00 (2018) 000–000 3
pressure on manufacturers to have to collect and analyze data, in order to stay competitive on the market, which is
expected to be pretty hard on small and medium enterprises (SMEs) due to their typically limited resources.
Nine main components of Industry 4.0 have been recognized as main drivers of its development and convergence
[15], i.e.: System Integration, Big Data and Analytics, Simulation and Virtualization, Internet of Things (IoT), The
Cloud, Cybersecurity, Autonomous Robots, Augmented Reality, and Additive Manufacturing. The core of these
nine components is the data and its processing to permit automation and improvement of manufacturing and other
related processes. Most of the components (6 out of 9) provide useful supports for collecting, storing and processing
manufacturing data, as illustrated in Fig. 1, where we recognize the nine main components of Industry 4.0 into
operations-related and data-related components. This further shows how important and central data is with respect to
the Industry 4.0.
Figure 1. The important components of Industry 4.0.
The main aims of the Industry 4.0 are to improve manufacturing efficiency and diminish the associated cost.
Measuring and enhancing reliability has a substantial impact on realizing these aims. The advantages of the new
industrial revolution, however, also bring along new challenges. The development of Industry 4.0 is expected to
effect important changes in the society. In Germany, for instance, as a leading manufacturing nation, they have
recognized four main areas that are going to be impacted by the Industry 4.0 [9], as follows:
• Productivity, by ca. 10-30% increase,
• Income growth, by ca. 1 percent of Germany’s GDP,
• Employment, by alteration in the requirement of skills, e.g. employees skilled in mechatronics and software
developments are predicted to have an improved demand, at the expense of low-skilled laborers, and
• Investment, by expecting investment of ca. EUR 250 billion during the next ten years in Germany in adapting
Industry 4.0.
Alike impacts are anticipated on a global scale [16], although at slightly different paces.
2.2. Reliability in Manufacturing Systems
Manufacturing systems, being cyber-physical systems, have three factors that affect their reliability: software
reliability, hardware reliability, and reliability due to human interaction. All three factors have distinctive natures,
and as such, need to be analyzed first individually, and then combined to provide a unifying measure of the overall
reliability.
694 Sanja Lazarova-Molnar et al. / Procedia Computer Science 151 (2019) 691–698
4 Author name / Procedia Computer Science 00 (2018) 000–000
As elaborated in [17], in Figure 2 we illustrate failure rates of software and hardware components [18]. Hardware
components have three phases: in the beginning while they are burning in, when the effect of manufacturing errors
occurs, then the failure rate stabilizes until the “wear out” period, when, again, there is an increased failure rate due
to the components’ aging. Software components have the highest failure rate at the beginning, during integration and
test. During this testing phase faults are being removed and the failure rate decreases and stabilizes, until the next
upgrade. In Figure 1b we can see that software components do not exhibit the last “wear out” phase when the failure
rate increases. Instead, the last phase with software is the Obsolescence, where the component becomes obsolete and
no upgrades are being provided. One of the biggest differences between software and hardware reliability is that
software reliability does not depend on time, whereas the hardware reliability does. On the other hand, software
reliability is highly dependent and sensitive to changes in the environment, such as updates of hardware of software
components, as well as other relevant events.
Figure 2. a) Bathtub curve for hardware reliability; b) Revised bathtub curve for software reliability [18]
Besides software and hardware-related reliability, we also need to consider the effect that human interaction has
on a system’s reliability as manufacturing systems are often operated by humans at certain levels. Human interaction
increases the likelihood that the manufacturing system is used or operated in an unintended or wrong way, which
would result in reduced reliability and greater vulnerability to faults. A large portion of manufacturing systems are
designed for intensive interaction with humans and, thus, humans have, to a large extent, influence on the operation
of these systems. There have been numerous research efforts directed at estimating the impact of unexpected and
unintended human interaction with a wide range of Cyber-Physical Systems (CPS). One example of CPS with high
level of human interaction are smart buildings, for which a lot of research has been done on the topic of occupants’
behavior [19, 20]. Humans are highly uncertain and difficult to predict in their behavior. Therefore, modeling
reliability-related human behavior needs different approaches from the ones applicable to software and hardware
aspects.
Therefore, to obtain comprehensive measure of the reliability of a manufacturing system, all three types of
reliabilities need to be considered, as well as their interplays and interdependencies. This is far from trivial to do,
however, the availability of the IoT infrastructure may make it easier to extract and model relevant events and
interdependencies accurately, such as to obtain accurate reliability measures.
3. Reliability Analysis in Industry 4.0 Context
With the development of the industrial Internet of Things (IoT), new opportunities for analyzing reliability of
systems are being developed, as well as opportunities to validate the existing approaches. Typically, a lot of expert
knowledge is utilized for reliability analysis, and to a large extent, expert knowledge will remain irreplaceable for
safety-critical systems, where there is no option for collecting data on faults, as they should not happen at all due to
Sanja Lazarova-Molnar et al. / Procedia Computer Science 151 (2019) 691–698 695
Author name / Procedia Computer Science 00 (2018) 000–000 5
their catastrophic consequences. One example, where a great level of expert knowledge is needed is the design of
Fault Trees for reliability of aviation systems [21]. However, large portions of systems utilized in manufacturing are
not safety-critical, i.e. their faults and failures do not cause damage to the people or the environment. Mainly,
consequences of faults and failures in manufacturing systems are in terms of financial cost.
Therefore, the recent ICT developments and their use in manufacturing facilities create a significant opportunity
to effectively gather data on faults and failures of these systems and utilize it to supplement expert knowledge and
build more accurate reliability models. The development of Industry 4.0 has yielded a number of new moments and
associated opportunities that can change the way in which reliability in manufacturing systems is analyzed and
assessed. In particular, we focus on the following aspects:
1. Availability and ease of collection of data,
2. Large portion of systems being non-safety-critical,
3. Same flexible machines being utilized by different manufacturers for different purposes, and
4. New technologies lead to more complex and failure-prone systems.
Each of the four listed aspects provides different benefits and challenges, detailed as follows.
The availability and ease of collection of data has yielded collections of new types of data, but also requested
development of new and sophisticated approaches to enable full benefit of the data. E.g. typically, the data collected
is in form of time series, without explicitly capturing faults occurrences. This implies that there is high necessity of
approaches that focus on event detection, such that faults and other events occurrences can be extracted from the
time series data. Furthermore, accurate root cause analysis methodology will be the next requirement; such as to
extracts events dependencies and model them. Once such approaches are sufficiently advanced, reliability analysis
of systems can be automated, such as to be automatically performed, based on data from manufacturing machines.
The fact that most of the manufacturing machines are not necessarily safety-critical, and faults/failures are relatively
common occurrences, makes the data-based approaches very adequate. Collection of data for reliability analysis in
safety-critical systems would be impossible, as in those systems the failures can cost human lives. Examples are
automotive systems or aircraft. This brings us to the second point that supports the first.
The third fact on our list is the anticipation for flexible machines that can perform different tasks, which implies
that highly repetitive use same types of machines will be occurring. Thus, a lot more data can be collected for those
machines, yielding more accurate models. However, as we state by the fourth aspect, the availability of data and the
new technologies comes at a cost, which is the increased complexity of the systems. This increased complexity
implies vulnerability to faults and failures. E.g., CPS’ processes often depend on input from sensor data, so a fault in
the sensors can lead to significant damaging and costly consequences. This vulnerability only emphasizes the need
for accurate and efficient reliability and overall health assessment of the manufacturing systems.
To summarize, in Figure 3, we illustrate the feedback loop that can enabled by the data-based reliability
assessment for smart factories. We begin from the bottom left corner of the figure, where we have the smart factory
that collects a data from its manufacturing processes. This data then needs to be processed in a way that key
reliability-related events are detected (such as fault occurrences, repair starting times and completion times, etc.) and
extracted [22, 23]. Next, this data is used for learning fault models, including causality among faults and failures,
which is followed by advanced simulation and data analytics. The results of simulation and data analytics are
utilized for decision support on improved system configuration and generation of preventive maintenance schedules
for increased reliability of the system (i.e. smart factory in this case).
It is evident that machine learning and simulation will play important role in making the data-based reliability
analysis processes possible. Advanced event detection methods will be crucial for gathering data for building
reliability-relevant models, and accurate and efficient simulation methods will be needed for evaluating the
reliability of the built models. The automatically generated models can then be used for analyzing alternative
configurations for given systems, with the purpose of optimizing their reliabilities.
The four described aspects that characterize the latest developments, captured in the Industry 4.0 initiative, are in
favor of using data-based approaches for reliability assessment of manufacturing cyber-physical systems. In the
following section, we discuss the implications of the data-based reliability assessment of manufacturing systems.
6696 Sanja Lazarova-Molnar
Author name / ProcediaetComputer
al. / Procedia Computer
Science Science
00 (2018) 151 (2019) 691–698
000–000
Figure 3. Data for supporting reliability modeling in smart
factories
4. Implications
Managers and engineers in different manufacturing industries continuously plan with intention to improve
productivity, quality, and reliability of their manufacturing processes and the quality of their products. At the same
time, they constantly attempt to cut down manufacturing expenses, improve resource utilizations, and increase
manufacturing safety. These objectives can be achieved well with leveraging the capabilities of Industry 4.0 for
manufacturing. As both IoT and CPS are used to enable connectivity among different manufacturing components
and smart cyber controls, more accurate data about different manufacturing processes can be collected, stored, and
analyzed. In addition to this data from manufacturing processes, other business data from different enterprise
systems such as Enterprise Resource Planning (ERP), Customer Relationship Management (CRM), Supply Chain
Management (SCM), and Product Life Cycle Management (PLM) can provide detailed information about all
production and related processes.
With this collected data, manufacturers can be armed with better data-driven decision making to significantly
improve their operations and profitability. Different advanced cyber techniques such as simulation, data mining,
process optimizations, and machine learning algorithms can be applied to this data to improve the reliability of
Sanja Lazarova-Molnar et al. / Procedia Computer Science 151 (2019) 691–698 697
Author name / Procedia Computer Science 00 (2018) 000–000 7
different manufacturing systems, processes, and their products. For example, the collected data can be utilized by a
number of data-driven decision-making processes to enhance reliability of different manufacturing processes with
an aim to reduce materials, resources, and energy used to produce different products.
In addition, the collected information from different systems and processes can be utilized to optimize
maintenance schedules and perform preventive diagnostics. Cost and time are two important aspects in
manufacturing systems [24]. Consequently, manufacturing systems’ maintenance is a tremendously important
process that allows to keep manufacturing machines in outstanding operating settings and avoid unexpected
accidents that may outcome in process interruptions and down-times as well as costly repairs. In addition,
maintenance and support costs are estimated to be around 60 to 75% of the total lifecycle costs [25]. To optimize
maintenance and decrease interruptions and repair costs, the collected information can be applied to form a
knowledgebase to store, organize and analyze information using smart algorithms to propose optimizations for the
maintenance schedules and assist in noticing potential problem areas before an actual problem happens.
Another significant opportunity is the potential for Collaborative Data Analytics among manufacturing facilities
[11], which presents an opportunity to faster build more accurate fault models of machines that partake in the
different flexible manufacturing systems. By joining data on matching machines, and thus sharing the conditions
under which their machines operate, manufacturers can easily detect optimal configurations for their machines, as
well as obtain accurate fault models for designing maintenance schedules that reduce the costs and downtime. This
service could be either provided by a third party, or by joint companies’ initiative, in the form of an interest group,
which will guarantee anonymity and protection of companies’ data. We are aware that many companies will be
initially skeptical about sharing data, but we believe the necessity for survival will have an impact in changing these
views and subscribing to the pragmatic initiatives.
5. Summary and Outlook
In this paper we examined the concept of reliability within the context of Industry 4.0, especially the impact of
the obtainability and availability of data has on reliability analysis and assessment. The relevance of providing
accurate and efficiently calculated reliability assessment of a manufacturing system has been high priority for a long
time. What has changed now is the availability of new Information and Communication Technologies that have the
potential to transform the traditional approaches, and enhance the quality of the provided assessments. The biggest
game changer is the ease of obtaining and processing data on various phenomena related to the reliability. In line
with the new advancements, we outlined fours aspects that can impact the ways of assessing reliability for
manufacturing systems within the Industry 4.0 developments. One challenge that we foresee for obtaining useful
reliability-related data is obtaining data on events, such as faults, failures, repairs, etc. This event data is typically
facilitated through the challenging and intrusive process of event logging, or through advanced data analytics
approaches in event detection, which again need to train on some data as well, so event logging to some extent
cannot be avoided. Therefore, we see the success of the new and automated approaches on reliability assessment of
manufacturing systems to a large extent dependent on the development of approaches for accurate event detection
and process mining.
References
[1] K. Miriyala, N. Viswanadham, Reliability analysis of flexible manufacturing systems, International Journal of Flexible Manufacturing
Systems, 2 (1989) 145-162.
[2] S. Chakraborty, B. Ankiah, Assessment of Manufacturing System Reliability: A Case Study, The Journal of the Operational Research Society,
40 (1989) 55-63.
[3] T. Xia, Y. Dong, L. Xiao, S. Du, E. Pan, L.J.R.E. Xi, S. Safety, Recent advances in prognostics and health management for advanced
manufacturing paradigms, (2018).
[4] M. Catelani, L. Ciani, M.J.I.S.J. Venzi, RBD Model-Based Approach for Reliability Assessment in Complex Systems, (2018) 1-9.
[5] L. Columbus, Industrial Analytics Based On Internet Of Things Will Revolutionize Manufacturing, in, Forbes, 2016.
698 Sanja Lazarova-Molnar et al. / Procedia Computer Science 151 (2019) 691–698
8 Author name / Procedia Computer Science 00 (2018) 000–000
[6] E. Chiodo, D. Lauria, Some basic properties of the failure rate of redundant reliability systems in industrial electronics applications, IEEE
Transactions on Industrial Electronics, 62 (2015) 5055-5062.
[7] K.L. Wong, What is wrong with the existing reliability prediction methods?, Quality and Reliability Engineering International, 6 (1990) 251-
257.
[8] H. Lasi, P. Fettke, H.-G. Kemper, T. Feld, M. Hoffmann, Industry 4.0, Business & Information Systems Engineering, 6 (2014) 239-242.
[9] M. Rüßmann, M. Lorenz, P. Gerbert, M. Waldner, J. Justus, P. Engel, M. Harnisch, Industry 4.0: The future of productivity and growth in
manufacturing industries, Boston Consulting Group, 9 (2015).
[10] N. Mohamed, J. Al-Jaroodi, S. Lazarova-Molnar, Leveraging the Capabilities of Industry 4.0 for Improving Energy Efficiency in Smart
Factories, IEEE Access, 7 (2019) 18008-18020.
[11] S. Lazarova-Molnar, N. Mohamed, J. Al-Jaroodi, Collaborative Data Analytics for Industry 4.0: Challenges, Opportunities and Models, in:
2018 Sixth International Conference on Enterprise Systems (ES), 2018, pp. 100-107.
[12] A.J.N.N. Kusiak, Smart manufacturing must embrace big data, 544 (2017) 23.
[13] F. Tao, Q. Qi, A. Liu, A.J.J.o.M.S. Kusiak, Data-driven smart manufacturing, 48 (2018) 157-169.
[14] G. Richards, Warehouse management: a complete guide to improving efficiency and minimizing costs in the modern warehouse, Kogan
Page Publishers, 2017.
[15] G.-J. Cheng, L.-T. Liu, X.-J. Qiang, Y. Liu, Industry 4.0 development and application of intelligent manufacturing, in: Information System
and Artificial Intelligence (ISAI), 2016 International Conference on, IEEE, 2016, pp. 407-410.
[16] H. Kagermann, R. Anderl, J. Gausemeier, G. Schuh, W. Wahlster, Industrie 4.0 in a Global Context: strategies for cooperating with
international partners, Herbert Utz Verlag, 2016.
[17] S. Lazarova-Molnar, N. Mohamed, H.R. Shaker, Reliability modeling of cyber-physical systems: A holistic overview and challenges, in:
2017 Workshop on Modeling and Simulation of Cyber-Physical Energy Systems (MSCPES), 2017, pp. 1-6.
[18] J. Pan, Software reliability, Dependable Embedded Systems, Carnegie Mellon University, 18 (1999) 1-14.
[19] S. Lazarova-Molnar, H.R. Shaker, A Conceptual Framework for Occupant-Centered Building Management Decision Support System, in:
12th International Conference on Intelligent Environments, IOS Press, London, United Kingdom, 2016, pp. 436-445.
[20] O. Masoso, L.J. Grobler, The dark side of occupants’ behaviour on building energy use, Energy and buildings, 42 (2010) 173-177.
[21] F. Netjasov, M. Janic, A review of research on risk and safety modelling in civil aviation, Journal of Air Transport Management, 14 (2008)
213-220.
[22] V. Chauhan, B.J.P.M. Surgenor, A comparative study of machine vision based methods for fault detection in an automated assembly
machine, 1 (2015) 416-428.
[23] A. Theorin, K. Bengtsson, J. Provost, M. Lieder, C. Johnsson, T. Lundholm, B.J.I.J.o.P.R. Lennartson, An event-driven manufacturing
information system architecture for Industry 4.0, 55 (2017) 1297-1311.
[24] G. Chryssolouris, Manufacturing systems: theory and practice, Springer Science & Business Media, 2013.
[25] S. Takata, F. Kirnura, F.J. van Houten, E. Westkamper, M. Shpitalni, D. Ceglarek, J. Lee, Maintenance: changing role in life cycle
management, CIRP Annals-Manufacturing Technology, 53 (2004) 643-655.
