École de Gestion
Industry 4.0: from strategic maturity modelsto operational deployment usinglean six sigmatools
par
Pablo Ernesto de Paiva Pereira
Mémoire présenté àla faculté d’administration en vue del’obtention du grade de
Maître en Sciences, M. Sc. Stratégie del’intelligence d’affaires
Juin 2018
SUMMARY
To remain competitive, companies must invest in emerging technologies that charac-terize theindustry 4.0 (I4.0). To make the shift to I4.0, managers need to define their strategic direction and then deploy it to the operational level. The scientific literature presentsfew adequateresponsestothis need,focusing either atthe strategiclevel or at the operationallevel separately. In particular,there are few approachesto managethe transformation at the operational level. Considering this gap in the literature, this re-search aims to develop a framework to guide managers in their decision-making to-wards digitaltransformation. Based onthe Design Science methodology,we first pe r-formed a systematicliteraturereviewtoidentify approachesfortheformulation ofI4.0 strategies based on maturity models, as well as sometools to work at the operational level. At this level, we opted to work mainly with Lean Six Sigma (LSS)-based tools to supporttheimplementation ofI4.0,likethe project charter andthe value stream map (VSM) ,sinceitis an established standardinthe manufacturing world,thusfacilitating thetransformation processfor operations managers. Next, a newframeworkis pro-posedfromthesetheoreticalfoundationscombining LSSand BusinessIntelligence concepts, which werelatertestedinarealcasestudyina manufacturingcompany. Results suggestthatthe proposed approach hasthe potentialto support managersinthe I4.0transformation process.
Key words:Industry 4.0,leansix-sigmatools,lean manufacturing, businessintell i-gence, strategy deployment
RESUMÉ
Pour demeurer compétitives,les entreprises doiventinvestir dansles techno-logies émergentes qui caractérisentl’industrie 4.0(I4.0). Pour apporterle changement versl’I4.0,les gestionnaires doivent définirles orientations stratégiques et puis la dé-ployer, au niveau opérationnel.Lalittérature scientifique présente peu deréponses adé-quates à cette problématique, se concentrant soit au niveau stratégique, soit au niveau opérationnel séparément. Considérant cette lacune dans la littérature, cette recherche visele développent d’un cadre conceptuel qui permet de guiderles gestionnaires dans leur prise de décision detransformation numérique. La méthodologieadoptéeaété fondée surle « design science », composé de 5 étapes. La première est constituée d’une revue systématique delalittérature qui a permis d’identifier des outils utilisés pourla formulation des stratégies de l’I4.0 basés sur les modèles de maturité. Aussi,il a été possible d’identifier des outils sur le plan opérationnel, notamment, certains outils du Lean Six Sigma (LSS),commele « project charter » etle « value stream map », pour soutenir la mise en œuvre de la I4.0, puisque le LSS c’est un standard établi dans le monde manufacturier, ce qui facilite le processus de transformation pour les respon-sables des opérations. Ensuite, un cadre conceptuel a été développé à partir de ces bases théoriques, en combinant des outils LSS et les concepts d´intelligence d’affaires, ap-pliqués dans un casréel auprès d’une entreprise manufacturière. Les résultats suggèrent que l’approche proposée a le potentiel de soutenir les gestionnaires dans le processus detransformation vers l’I4.0
TABLE OF CONTENTS SUMMARY... 3 RESUMÉ... 4 TABLE OF CONTENTS ... 5 LIST OF TABLES ... 8 LIST OF FIGURES...9 LIST OF ACRONYMS ... 10 ACKNOWLEDGMENTS... 11 1 INTRODUCTION... 12 2 BACKGROUND... 17
2.1 Industry 4.0 principles... 17
2.1.1 Industry 4.0 strategy framework ... 17
2.2 Maturity model andreadiness definition ... 18
2.3 IMPULS – Industry 4.0 readiness... 19
2.3.1 IMPULS – Smart operations...21
2.3.2 IMPULS - Smart factory...22
2.3.3 IMPULS – Strategy and organization...24
2.3.4 IMPULS – Smart products... 24
2.3.5 IMPULS – Data driven services...24
2.3.6 IMPULS – Employees...24
2.4 Lean Six Sigma (LSS)...25
2.4.1 Lean Manufacturing (LM)...25
2.5 Background analysis... 26
3 METHODOLOGY... 27
3.1 Design Science...27
3.1.1 Step 1 – Systematic Literature Review...29
3.1.3 Step 3 – Applyingframework developed for data collection (proo f-of-concept) 31
3.1.4 Step 4 – Evaluation ofthe conceptual framework...32
3.1.5 Step 5 – Improvethe conceptualframework...33
4 Systematic Literature Review Results...34
4.1 Bibliometric results...34
4.2 Content results...36
4.3 Literature review discussion...40
4.4 Research gapsidentifiedintheliterature...44
5 THEORETICAL FRAMEWORK: LSS ADAPTED TOOLS ... 45
5.1 I4.0 Envision – IMPULS diagnosticaudit... 46
5.2 I4.0 Enable – Project Charter...46
5.3 Enact - LSS adaptedtoolsto I4.0...47
5.3.1 Classic VSM...47
5.3.2 VSM adaptedto I4.0 – VSM 4.0...49
6 APPLICATION - PROOF-OF-CONCEPT CASE...54
6.1 Company Profile...54
6.2 Strategical Level - Envision - IMPULS as-is state...54
6.3 Enable - Project charter as-is state...57
6.4 Enact - VSM 4.0 as-is state...59
6.5 VSM 4.0future State...66
7 DISCUSSION... 72
7.1 Conceptual model evaluation...74
7.1.1 Conceptual model evaluation – Researcher point of view...74
7.1.2 Conceptual model evaluation – User’s point of view...75
8 CONCLUSION... 76
8.1 Academic contributions...76
8.2 Managerial contribution...77
8.4 Avenues of research...78
9 REFERENCES... 79
ETHIC COMMITTEE APPROVAL...92
LIST OF TABLES
Table 1: IMPULS dimensions and related fields... 21
Table 2: Design science steps... 28
Table 3: Input phase forliterature review(research protocol)... 29
Table 4: Systematic Literature review results... 30
Table 5: TAM criteria... 33
Table 6: I4.0MM content Analysis... 37
Table 7: I4.0 & LMintegration summary... 39
Table 8: IMPULS dimensions and I4.0 principles... 45
Table 9 - IMPULS evaluation by dimension, adapted from Lichtblau et al. (2015). 51 Table 10: BIform – Fields description... 52
Table 11: Data quality evaluation... 52
Table 12: Company profile – Case Study... 54
Table 13: IMPULS results – as-is state... 55
LIST OF FIGURES
Figure 1 -I4.0 Digitization process dimensions, adapted from Lichtblau et al. (2015)
... 14
Figure 2 – IMPULS dimensions, Lichtblau et al. (2015) ...20
Figure 3 - Articles by publication years... 34
Figure 4 - Articles by periodictype... 35
Figure 5 - Articles by country of the principal author... 35
Figure 6 - Articles by research strategy... 36
Figure 7 - Articles by I4.0 & LMintegrationtype... 38
Figure 8 -I4.0 strategy framework. Adapted from Erol et al. (2016) ... 46
Figure 9 -I4.0 project charter, adapted from Pyzdek et al. (2010) ... 47
Figure 10 - VSM 4.0 – Adapted from Meudt et al. (2017) ... 50
Figure 11 - BI process, Smart factory, adapted from Meudtet al. (2017)... 53
Figure 12 - Project charter first version, adapted from Pyzdek et al.(2010)... 58
Figure 13 - VSM 4.0 as-is state (adapted from Meudt et al. (2017))... 63
Figure 14 - BI as-is state, machining +inspection, adapted from Meudt et al. (2017) ... 64
Figure 15 - BI as-is state, assembly prep. + assembly (adapted from Meudt et al., 2017)...65
Figure 16 - VSM 4.0 future state, adapted from Meudt et al. (2017) ... 69
Figure 17 - BI future state, machining, adaptedfrom Meudtet al. (2017) ... 70
LIST OF ACRONYMS BI Businessintelligence
DMAIC Define, measure, analyse,improve FMEA Failure mode and effect analysis IIoT Industrialinternet ofthings I4.0 Industry 4.0
I4MM Industry 4.0 maturity model IoT Internet ofthings
LM Lean Manufacturing LSS Lean Six Sigma
M2M Machineto machine communications
MESI Ministry of economy, science andinnovation MRP Materials resource planning
OEE Overall equipment efficiency PLC Programmablelogic controller RFID Radio frequencyidentification SLR Systematicliteraturereview SME Small and medium enterprises SS Six sigma
TAM Technology acceptance model TPS Toyota production system VSM Value stream mapping
ACKNOWLEDGMENTS
I wouldliketo express my sincere gratitudeto Professors Luis Antonio de Santa Eulalia and Elaine Mosconi for their inspiration and dedication that made it possible forthis projectto berealized. I would alsoliketothankthe case study participants for their valuable contributiontothis project. I also extend my appreciationto allthe pro-fessors fromthe MSc Business administration, Business Intelligence, from Université de Sherbrooke,that madethisjourney possible.
Finally, I wouldliketothank my beloved wife and daughters, for being by my sideinthis adventure oflivingin another country.I would alsoliketothank my parents and sister,fortheir solid support, eventhoughthey werethousands of kilometres away.
“Look up at the stars and not down at your feet. Try to make sense of what you see, and wonder about what makesthe universe exist. Be curious” – S. Hawking
1 INTRODUCTION
The first wave of industrial revolution (Industry 1.0) occurred atthe end of the 18th century, withthefirst automated machines, via steam or water power. Industry 2.0 happened around 1870 withthe concept of assembly lines, helped by electricity. Industry 3.0 beganin 1969, withthefirst programmablelogic controller(PLC) andthe use of electronics and information technology in production automation (Deloitte, 2015).
The concept of Industry 4.0 (I4.0) emergedinthe mechanicalindustry of Ger-manyin 2011(Lichtblau et al., 2015). Inthe United States,theterm usedistheIndus -trial Internet of Things (IIoT), whichinvolves not only manufacturing, but also other industries, such as health care, civil construction, etc. Duetoitsimportance atthe in-ternationallevel, I4.0 wasthe subjectofthe World Economic Forumin 2016(theme: The FourthIndustrial Revolution, whatit means, howto respond) (Schuh et al., 2017). In Quebec, to deal with this phenomenon, the Government launched its "digital agenda",in 2016, andthereis a ministryresponsible forthe digitization oftheindus -tries,the Ministry of Economy, Science andInnovation (MESI). Animportant program was also launched by the MESI in 2017, the “Manufacturier innovant” (in French), which encompasses high investments in the manufacturing sector to help companies quicklyintegrate emergingtechnologies from I4.0.
I4.0 can be defined as a phenomenonin which emergingtechnologies of phys -ical, biological and digital worlds convergetorevolutionizethe organization of wor ld-wide value chains, changing business models, production, distribution and consump-tion (Schwab, 2016). I4.0 is a phenomenon that is part of the Fourth Industrial Revolutiontransformations (Schwab, 2016). Another definitionis areal-time network of people, equipment and objects used for business process management and for the creation of a value network (Dombrowski, Richter & Krenkel, 2017). It can also be
defined as a “collective term for technologies and concepts of value chain organ iza-tion”(Hermann, Pentek& Otto, 2016).Concerningtheinternal point of view ofindus -try, or vertical perspective,it consistsin theintegration of industry’s physical objects with the Internet, called the digital world. If we consider a horizontal perspective, it relatesto theapplication ofthesetechniquesthroughoutthe wholecompanysupply chain, as concernstheintegration of data and objects fromindustry suppliers,logistics partners, service providers and customers (Deloitte, 2015).
Inthe manufacturing environment,the digitization concept can be unfoldedin several dimensions, accordingto Figure 1, such asthe smart factory, where machines can be equipped with sensors so that they can communicate with the company in tra-net/internet and also communicate with each other (machine to machine commun ica-tion -M2M). Another dimension is smart operations, which consists of digital infor-mation sharinginternally and externally with the company, usage of cloudto store and analyze information, autonomous processes that self-react to the production envi ron-ment (Lichtblau et al., 2015). Thethird dimensionis smart products, which stands for installingsensorsinthe productthatallow datacollectionincustomer operations, providing new opportunitiesforthe companyto offer servicesintegratedtothe product, suchas preventive maintenance, product usageanalysis(Kolberg, D., Zühlke, D., Zuehlke, D., 2015). Finally, the fourth dimension is data-driven services, enabled by the smart productsinfrastructure, wherethe company can changeits business models, creating valuethrough product data analysis, adding digitalservicesto customers (Rymaszewska, Helo & Gunasekaran, 2015).
Figure 1 -I4.0 Digitization process dimensions, adapted from Lichtblau et al.(2015)
Research problem
In an I4.0 strategy transformation, there are breaking changes (worthy of an industrialrevolution) and also continuousimprovements. To makethese changes, com-panies mustfirstidentifytheir strategic orientation andthen deployittothe operational level. It is fundamental to make this strategic orientation of I4.0 so that the company canidentifyits current and desired situationtowardsI4.0. This strategy orientation can lead to a change in the business model, adding data-driven services that can provide newsources ofrevenue. Theaccess of a large quantity of datafromsmart products makes value creating possible while generating profit(Rymaszewska et al., 2015).I4.0 can enable the application of the Product Service System, “where companies develop products with value-added services,instead of a single productitself and providetheir customers with needed services (Lee, Kao & Yang, 2014).Moreover,the connection betweenI4.0 strategy andthe operationallevel will provide company management with importantinformation, such asthe productivity gains afterI4.0implementation, quality improvement and soforth.
DIGITIZATION SMART
FACTORY
SMART
PRODUCTS DATA-DRIVENSERVICES SMART
OPERATIONS SERVICES
Itis at the operational levelthat gaps have been identifiedintheliterature, since there are just a few frameworks at the strategic level and no framework making explicit connectiontothe operationallevel. Atthe strategiclevel, we find mainlytools for evaluating maturitylevels, such astheIMPULSframework("impulse"in German) (Lichtblau et al., 2015). IMPULS is a maturity model, which is a collection of o rga-nized elementsthat definethe characteristics of effective processes at various stages of development (Pullen, 2007). Another exampleisthe conceptual framework of Erol et al.toformulatethestrategy ofcompaniestowardsI4.0(Erol, Schumacher & Sihn, 2016).
Asforthe operationallevel,the literatureislimited, butsomesuggestions exist on how to deploy strategy at the operational level. At this level, there are some works about Lean Manufacturing (LM) and Six Sigma (SS), presented as a possible solution forthis problem (Meudt, Metternich & Abele, 2017). LM isintended to p ro-duce products and services atlowest cost and atthetime required bythe client, while eliminating/reducing waste in the process. Six Sigma aims to reduce the variation of the process, using the approach DMAIC (Define, Measure, Analyze, Improve, Con-trol). Amongthe various methods of process continuousimprovement, SS and LM are considered amongthe best methodologies, widely usedin variousindustries. They are currently designated as the state of the art of continuous improvement (Salah, Rahim & Carretero, 2010).
However, to the best of our knowledge,two major gaps exist: no framework exists to connect the strategiclevel to the operational level; and the proposedtools at the operational level do not respond to the data-driven needs for smart factories and smart operations. Thus,this work aimsto contributetoreducing this gap andits ob jec-tives are statedinthefollowing.
Research objective
This research hasthe following objectives:
1. identifythe LSStoolsthat can helptotranslatethe strategic I4.0 objectives at the operationallevel;
2. checkifthese LSStools need adaptationtothe context of I4.0;
3. propose adjustments consistent with an I4.0 migration strategy, as appropriate; 4. assessthe contribution ofthe LSStoolsin an organizational context oftrans
for-mationtowards I4.0.
In orderto achievethe overall objectives,theresearch questionthatthis study seeksto answeris: What wouldthe roadmap betotranslatetheI4.0 strategyinto con-crete projects onthe shopfloor, for companiesthat arein anI4.0transformation p ro-cess?
2 BACKGROUND
Inthis chapter, we presentthe fundamental concepts ofthisresearch. 2.1 Industry 4.0 principles
Accordingtotheliterature,the principles of I4.0 are:interconnection, infor-mationtransparency, decentralized decisions,technical assistance. The principle of in-terconnection is the connection of machines, equipment, sensors, operators between them, but alsothe connection of alltheseelementstothe cloud. These communications standards and cybersecurity are also part of this principle. Information transparency contains analyses of the data andthe provision ofinformationto users. Itis a virtual copy of the physical world, createdthroughthe binding of data sensors with factory digitized models. Decentralized decisions allowthe use of computers, sensors and op-eratorsto monitor and controlthe physical world,in an autonomous manner. Technical assistanceisthe automation of physical and virtual operations (Hermann et al., 2016). Asthe definition of I4.0is complex, and may contain different definitions depending onthe stages ofthe company onthistopic, several models of maturity have been cre-atedto help companiesto positionthemselves onthistheme.
2.1.1 Industry 4.0 strategyframework
A framework concerning I4.0 vision and strategy building was proposed by (Erol, Schumacher & Sihn, 2016). Its main goal wasto help companies developtheir I4.0 objectives and clearly communicatethemtoits stakeholders. It consists of three phases, as described below:
Envision – This phase concernsthe company understanding of I4.0 concepts and the alignment of these concepts with company-specific objectives and customer needs. The stakeholders forthis phase are companytop management, and mayinvolve
otherimportant business partners and middle management, customers and external ex-perts ofI4.0.
Enable – This phaseturnsthe vision definedinthe envision phaseinto con-crete actions,transformingitinto a strategic plan; soit describes what hasto be done to achievethese objectives. These concrete actions are representedin fourlayers: Cus -tomer segments (market perspective, value proposition (product perspective), key re-sources, technologies and activities (process perspective) and the necessary partners (the network perspective).
Enact - This phasetransformsthe strategic plan defined atthe Enable phase into concrete projects, with atiming chart, definedteams and resources.
2.2 Maturity model and readiness definition
The word maturity has the following definitions: “state of being complete, perfect, or ready” and indicates some advancement in the development of a system. Therefore, maturing systems (e.g. biological, organizational or technological) raise their abilities over time concerningthe accomplishment of some wanted future state (Schumacher, Erol & Sihn, 2016).
Accordingto Wendler, (2012) “a clear definition oftheterm maturity model is often avoided. Publications of maturity models rather use descriptions of purpose and functioning of the models”. Maturity models generally consist of a sequence of levels (or stages)thatform a projected, wanted, orlogicaltrack from aninitial state of maturity. Maturity models are usedto evaluate as-is situations,to guideimprovement initiatives, and to control evolution (Maximilian Röglinger, Jens Pöppelbuß, Jörg Becker, Maximilian Roeglinger, Jens Poeppelbuss, 2012). Maturity models proposeto
organizations a simple but effective opportunity to evaluate the quality of their p ro-cesses. Developed out of software engineering, the application fields have expanded and maturity model researchis gaining moreimportance (Wendler, 2012).
Maturity models definethe progress of an entity overtime. This entity can be anything ofinterest: a human being, an organizational function, etc. (Gabor, 2001). A maturity modelis an organized collection of elementsthat definesthe characteristics of effective processes at diverse stages of development. It also proposes points of sep-aration between stages and methods of transitioning from one stage to another’’ (Pullen, 2007). ‘‘A maturity model consists of a sequence of maturitylevels for a class of objects. It represents an anticipated, desired, ortypical evolution path ofthese ob-jects shaped as discrete stages. Typically, these objects are organizations or pro-cesses.’’ (Becker, Knackstedt & Pöppelbuß, 2009).
A synonym for maturity modelis readiness models withthe aim of capturing the starting point and preparingthe development process. Readiness assessment occurs before engaginginthe maturing process. Inthe production area,recent readiness and maturity models have been used for example in eco-design, and utility management energy manufacturing orlean manufacturing (Schumacher et al., 2016).
2.3 IMPULS – Industry 4.0 readiness
This model was created bythe Germaninstitutions: VDMA, RWTH Aachen, IW Consult. The design ofthis maturity model was performed using a mixed me thod-ology of an analysis oftheliterature, expertise, workshops, and a comprehensive com-pany survey. The study defined six readinesslevels: 0-Outsider; 1- Beginner; 2- In ter-mediate; 3- Experienced; 4- Expert; 5 -Top performers. The Readiness Model is founded onthe four dimensions ofIndustry 4.0 (Smart factory, Smart operations, Smart products, Data-driven services. The model identifiedtwo additional, commonly appro-priated dimensions that were also taken into account: strategy and organization, and
employees. Allin all,the modelthenlooks at six dimensions: Strategy and organ iza-tion, Smart factory, Smart operations, Smart products, Data-driven services and Em-ployees. Figure 2 delivers an outline ofthe structure ofthe Readiness Model. It shows the six basic dimensions. Thetable 1 showsthe fields related with each ofthe six d i-mensions (Lichtblau et al., 2015).
Figure 2 – IMPULS dimensions, Lichtblau et al. (2015) IMPULS
Strategy and organization
Smart factory Smart operations Smart products Data-driven services Employees
Table 1: IMPULS dimensions and related fields IMPULS Dimensions IMPULS Fields
Strategy and organization Strategy Investments
Innovation management Smart Factory Digital modelling
Equipmentinfrastructure Data usage
IT Systems Smart operations Cloud usage
IT security
Autonomous processes Information sharing
Smart products Data analyticsin usage phase ICT add-on functionalities Data-driven services Data-driven services
Share of revenues Share of data used Employees Skillacquisition
Employee skill sets Lichtblau et al.(2015)
2.3.1 IMPULS – Smart operations
Smart operations concerns the degree of autonomous processes, the degree of in for-mation sharing with other processes (Horizontal and Vertical Integration) and IT secu-rity and cloud usage. Horizontal meanstheintegration of allinternal and external actors in a value chain, from suppliers,internal production,to customers. Verticalintegration stands forintegration onlyinsidethe company,from product developmentto planning, production, after sales, finance, marketing, etc. (Lichtblau et al., 2015). Autonomous
processes stand forthe workpieces moving bythemselvestothe next processing s ta-tion, the establishing of process sequences by their own, and the communication of production parameterstothe equipment (Lichtblau et al., 2015). “Autonomyin general means the independence of a system in making decisions by itself without external instructions and performing actions byitself without external forces”. “Autonomy of a system meanstwo basic characteristics: First, independence from neighbour systems and fromits environment, and second,the abilityto controlitself”. The combination of autonomous resources (cells, robots,transport systems) with autonomous parts, subas -semblies and products will drive to autonomous processes (Scholz-Reiter & Freitag, 2007).
2.3.2 IMPULS - Smartfactory
Regardingthe smart factory, the equipmentinfrastructure checksifthe pieces of equipment are ableto communicate with each other,i.e. machineto machine com-munication (M2M), areinteroperable and can be controlled. M2M happens between machines (some objects or devices) with computing/communication capabilities wi th-out humanintervention. M2M usesthe machinesto monitor certain events with sensors andtoinstruct actuation. The captured data are relayedthrough wired or wireless ne t-worksto servers, which extract and processtheinformation gathered and automatically control andinstruct other machines (Kim, Lee, Kim & Yun, 2014). Thelogic concern-ing M2M communicationsis based onthree factors: 1) a networked machineis more valuable than an isolated one, 2) when multiple machines are interconnected, more autonomous applications can be achieved, and 3) smart and ubiquitous services can be enabled by machine-type devicesintelligently communicating with other devices any-time and anywhere(Chen, Wan, Gonzalez, Liao & Leung, 2014).
2.3.2.1 Data Collection and Usage (Businessintelligence concepts)
The smart factory presents the items of data collection and data usage. To explainthetheorythat supportsthese dataitems, we can refertothe Business Intell i-gence (BI) concept. The BI can be defined as: “A combination of processes, policies, culture, andtechnologiesfor gathering, manipulating, storing, and analyzing data co l-lected frominternal and external sources,in orderto communicateinformation, create knowledge, andinform decision-making” (Foley & Guillemette, 2010).
Accordingto Foley et al.(2010)the BI can be viewed as a process. The first phaseis data collection, wherethetype of datato be collectedisidentified, for example, the setup time or manufacturingtime, and the frequency of this data collection. The second phaseis data storage, which could be in a digital (relational database, spread-sheets, data warehouses) or paper form. Afterthe datais storedthere should be a critical analysis of its quality. Data quality has several dimensions, such as accuracy, timel i-ness, precision, reliability, currency, completeness, and relevancy (Wang & Strong, 1996). The third phase is data visualization, which could be in the form of reports and/or dashboards. Dashboards are visual and interactivetoolsthat allow usersto watch relevant performanceindicators for the company. Dashboardinformation helps users intheir decision processto achieve company goals (Maddah, 2013). The fourth phase is analysis, which stands for applying statistical methods and/or computationaltoolsto discover relevant patternsin the business. The analysis can be descriptive, predictive and prescriptive. Descriptive analytics finds patterns and relationshipsin historical and existing data ( Haas, Selinger & Tan, 2011). Predictive analytics embraces a diversity oftechniques, such as regression, neural networks, etc.,that predictfuture results based on historical and current data (Gandomi & Haider, 2015). Thefinal phaseis prescr ip-tive analytics,that has a “whatif” capability,it suggests actionsto different scenarios (Chae & Olson, 2013)
2.3.3 IMPULS – Strategy and organization
Strategy and organization is organized in three sub dimensions: Strategy, in-vestments, andinnovation management. Strategy verifiesif thereis a strategyto im-plement I4.0, andifit´s embeddedinthe overall strategy ofthe company. Ifthereis an I4.0 strategy,it should have a system ofindicatorstotrackit, andthe strategy should be revised in a defined frequency. Investments stand for planned investments that should be madeinthe company, sothatit can progressinthe I4.0transformation. In-novation management specifies that the company has to innovate in its manage-ment/business practices, enabling new business models, such as new services incorpo-ratedinits products (Lichtblau et al., 2015)
2.3.4 IMPULS – Smart products
This dimensionrelatestothetechnologiesincorporatedintothe products, such as sensors,that will allowthe companyto analyze product data onthe field, opening possibilitiesto provide new servicesto customers, and alsoto have predictive models, in ordertoreducethe product failure rate onthefield.
2.3.5 IMPULS – Data driven services
The smart productstechnologies allow companiesto offer new servicestoits customers, based upon data analysis coming from product usage. This dimension checksif data-driven services are available atthe company,the percentage ofthe rev-enues generated bythem, andif this datais sharedthroughoutthe company, and with its customers.
2.3.6 IMPULS – Employees
I4.0 presents new challenges to employees, because they must acquire new qualificationsin orderto be ableto work withits newtechnologies and processes. This
dimension allows checking whetherthereis a definition ofI4.0 skills, andifthese skills are assessed andimplemented.
2.4 Lean Six Sigma(LSS)
LSS is the integration between the methodologies of Lean Manufacturing (LM) and Six Sigma (SS). LMisintendedto produce products and services atlowest cost and at the time required by the client, while eliminating/reducing waste in the process. SSis aimed at reducingthe variation ofthe process, usingthe DMAIC (Define, Measure, Analyze, Improve, Control) approach. Among the various methods of process continuous improvement, SS and LM are considered among the best methodologies, widely usedin variousindustries. They are now known asthe state of the art of continuousimprovement (Salah, Rahim & Carretero, 2010)
2.4.1 Lean Manufacturing (LM)
The aim of (LM)isto be very responsiveto customer demand via waste reduc-tion. The LM goal is to produce products and services at lowest cost and at a time required by the customer. The lean concept was created in Japan after the Second World War when Japanese manufacturers understoodtheirlack of capacitytoinvestin the rebuilding ofthe damaged facilities. Japanese car manufacturers,like Toyota, pro-duced cars with fewer resources, such as: inventory, human effort, investment, and defects andintroduced a greater and ever-growing variety of products. LM gives com-panies a competitive advantage via cost reduction, productivity improvement and qua l-ity (Bhamu & Singh Sangwan, 2014).
The term LM was created in the International Motor Vehicle Programme by researchers ofthe Massachusetts Institute of Technology, withthe goal of unders tand-ingthe performance gap between American and Japanese car manufacturers. Womack,
Jones & Roos (1990) defines LM as a dynamic process of change, driven by a sys tem-atic set of principles and best practices aimed at continuousimprovement. LM com-binesthe best features of both mass and craft production.
The Lean Manufacturing concept was derived from Toyota Production System (TPS). The birth of lean was in Japan within Toyota in the 1940s: The TPS was groundedinthe wishto producein a continuous flow, which wasthe opposite of along production runs system;it was based around the acknowledgementthat only a small part ofthetotaltime and effortto process a product added valuetothe end customer. This was clearlythe opposite of whatthe Western world was doing mass production based on material resource planning (MRP) and complex computing systemsthat were developing alongside mass production philosophies from Henry Ford,i.e. high volume production of standardized products with minimal product switches (Melton, 2005)
Sugtogmori et al.(1977) definedthe TPS with 2 main components: 1-" reduc-tion of cost through elimination of waste". This includes building a system that will systematically eliminate waste by consideringthat anything otherthanthe minimum amount of equipment, materials, parts, and workers which are undeniably essentialto production are simply surplus that only raises the cost; 2- " to make full use of the workers’ capabilities", which meansto consider andincludethe employeesinthe pro-cess ofimprovement.
2.5 Background analysis
Throughthese background analyses weidentifiedthe potential ofintegration between I4.0 and LSS, and a systematic literature review will later be performed to explorethis potential, and will be explainedin Chapter 3, methodology.
3 METHODOLOGY
In orderto attain our research objectives,the research methodology adopted was based on 5 phases, based onthe design science approach, which will be explainedin Section 3.1
3.1 Design Science
This research has an exploratory nature andislimitedtothe use ofthetoolsin the first phases ofthe LSS: Define and Measure. Sincethe workisintendedto create or adapt a fewtoolsto deploythe strategy at the operationallevel, we have adopted a Design science approach (Hevner, March, Park & Ram, 2004) with a proof-of-concept to be realizedin a manufacturing companyin Quebec.
The Design science approach was chosen because it helpsto develop andtest the proposed approach. Design Science offers guidelinesto ensurethe quality of the research,including:the development of a conceptual framework,the definition ofthe relevance ofthe problem andthe contribution ofthe research; verification ofthe rigor of the research, realizing several iterations and communicating the results of the re-search (Hevner et al., 2004).
Design science classifiesits artifactsin4 categories: constructs, model, method, instantiation. Constructs form the vocabulary of a domain, and are used to describe problems andtheir solutions. A modelis a set of propositions or statements showing relationship among constructs. A methodis a set of steps (an algorithm or guideline) usedto perform atask. Instantiation concerns the operationalization of constructs, mod-els, and methods(March & Smith, 1995).Inthis research,the artifacts are classified as model, because ofits framework adaptationto I4.0, and also are classified as method, becausethereis a step by stepinstructionto deploy I4.0 strategy.
The phases ofthis research project are: 1)literature review; (2) development ofthe conceptual framework; (3) application ofthe conceptual frameworkin a manu-facturingcompanyin Quebec;(4)assessment oftheconceptualframework;(5) im-provingthe conceptualframework. These phases are describedin Table 2 below.
Table 2 :Design science steps Design sc
i-ence steps Description Main expected results (1) deve
lop-ment oftheli t-erature review
Identify the theoretical frameworks to drivethis research, as well asthetools of the LSS andthe maturity oftheI4.0 mod-els
The systematic literature review is conductedtoidentifythe existing the-oreticalframework, LSStools and ma-turity models.
(2) deve lop-ment ofthe conceptual framework
Development ofthe conceptual
frame-work ofthe research anditstools Concepas a roadmaptual frameworkto deploy deve I4.0lopmen strategyt
(3) application ofthe concep-tual frame-work
The main data collectiontakesthe form of a mixed qualitative and quantitative studyin a manufacturing companyin Quebec
1 - The assessment ofthe maturity of the companyintheI4.0, usingthe questionnaire ofthe model "IMPULS" 2-Presentation ofthe resultsto man-agement and understanding ofthe strategy ofthe companytowardsthe I4.0 - 3 Evaluation of a production process,inidentifyingits current and future state usingthe LSStools (4) evaluation
ofthe concep-tual frame-work
The usefulness, quality and effectiveness ofthe conceptual framework/tools must be demonstratedrigorously,throughthe good execution ofthe assessment me th-ods.
Inthis regard, we usedthe approach of TAM ('Technology Acceptance Model')(Davis, 1985), which allows to assessthe usefulness and ease of use ofthe conceptual framework and tools.
(5)improve the conceptual framework
The results ofthetests must allowthe improvement ofthe model.
Data collected aboutthe application ofthetest and feedback fromthe company will beinputstoimprove the framework andtools.
This project provides only onei tera-tion because ofthe restriction oftime in a master's degree; However, newi t-erations are proposed.
3.1.1 Step 1 – Systematic Literature Review
This review was based onthe methodology designed by (Conforto, Amaral & Silva, 2011), comprisingthree phases.
3.1.1.1 Phase 1: Inputs
This phase consists ofthe definition oftheinputsfortheliterature review, as well asits planning. Table 3 below showsthe phase 1 protocol.
Table 3: Input phase forliterature review(research protocol)
Phase I4.0 Maturity models I4.0integration with LM Objectives Identifythe existing I4.0 maturity
models Idenintegratify tionthe w currenith LMt works about I4.0 Database
definition
ABI /INFORM, SCOPUS, SCIENCE DIRECT. Searches were done using « title », »ab-stract », »keywords »
ABI/INFORM, SCOPUS, SCIENCE DIRECT. Searches were done using «title », »abstract », »keywords » Search strings ((“Industr* 4.0” OR “Smart
Man-ufacturing” OR “Industrial Inte r-net of Things”) AND (“Maturity Models” OR “Readiness”))
((“Industr* 4.0” OR “Smart Manufac -turing” OR “Industrial Internet of Things”) AND (“Lean Manufac tur-ing” OR “Lean Production”)) Inclusion cr
i-teria
Maturity models that presents its dimensions, sub-dimensions, stages and questions. Articles in Englishlanguage only
Works that presents practical exam-ples of integration, linking I4.0 pr in-ciples/methods with LM princ i-ples/methods. Articles in English language only.
Qualification criteria
Type of Industry, Literature Type, differentiation aspects from other models, Implementation Data
Integration type, implementation data,tools forintegration
Filters First Filter (Reading title, ab-stract, key words); Second filter (Reading ofintroduction, conc lu-sion); Third filter (Complete read-ing)
First Filter (Reading title, abstract, key words); Second filter (Reading of introduction, conclusion); Third filter (Complete reading)
3.1.1.2 Phase 2: Processing
The results after performing allthe steps of phase 1. Data Processingis shown in Table 4. Concerningthe I4.0 maturity model,the search stringsidentified 30 articles (database extractiontotal). After duplicates removal,the results decreasedto 28. After the first and second filter,the number of articles droppedto 9 and 5. Finally, afterthe thirdfilter,there were 5 remaining articles. ConcerningI4.0integration with LM,the search strings identified 64 articles (database extraction total). After duplicates re-moval,theresults decreasedto 62. Afterthefirst and second filter,the number of art i-cles droppedto 31 and 17. Finally, afterthethirdfilter,there were 11remaining art i-cles.
Table 4: Systematic Literaturereview results
Filtering Phase I4.0 Maturity model
Number of articles I4tion.0 withintegra- LM
ABI/Inform 15 19
Science Direct 3 2
Scopus 12 43
Database extractiontotal 30 64
Duplicate removal 28 62
Filter 1: Abstract, Key words,title, references 9 31 Filter 2: Reading ofintroduction, conclusion 7 17
Filter 3: Complete reading 5 11
3.1.1.3 Phase 3: Outputs
The outputs are displayedinitem 4.2. The articles were classified for a better understanding ofthe research subject.
3.1.2 Step 2 – Framework development
The framework was chosen based onthe systematicliterature review (SLR) and will be adaptedif necessary sothatit can be used as aroadmapforI4.0 strategy deployment. The LSS operationaltools was also chosen based onthe SLR and adapted tothe I4.0 context and Business Intelligence process.
3.1.3 Step 3 – Applyingframework developedfor data collection (proof-of-concept) This step will definethe criteriato choosethe company forthe application of the framework and also describes strategical and operational data collectionthat was based ontwo specialtoolsidentified byliteraturereview step.
The company wheretheframework will be applied will bein anI4.0trans for-mation process and also will have used Lean Manufacturing methods andtoolsinits production process. This I4.0transformation process shouldinvolveinitialinvestments inthe fields of smart factory and smart strategicallevel data collection. The personthat will beinterviewed should bethe person responsible for operations.
This data collection ofthe strategy part will begin with the aid of IMPULS method, wherethe questionnaire of Annex 1 will be usedtointerviewthe directors of the companyin person. TheIMPULS method was chosen becauseitisthe only I4.0 maturity modelthat has a significant number ofimplementationsinthe manufacturing industry (see Table 3 ofthis document).. Afterthisinterview, a report of companies´ current I4.0 maturitylevel was generated withthe help of IMPULS onlinetool. Al t-hough the report shows the maturity level of all 6 IMPULS maturity levels, this re-search will prioritize only 2 IMPULS dimensions, smart factory and smart operations, forthe following reasons: 1)time restrictionin a Master’s degree research; 2) These dimensions arethe ones morerelatedtothe operationallevel ofthe company. Afterthe analysis of IMPULS report, a discussion of the company I4.0 strategy in the short, mediumtermis made withthe help ofthe form project charter. This discussion prior i-tizedthe company’s strategy concerningthe smartfactory and smart operations, where some high-level actions ofthese strategies were documented withthe desired comp le-tion dates
3.1.3.1 Operational Data collection
Concerning the operational part, the proposed VSM 4.0 tool (which will be presentedlater) was applied to a productionline ofthe company. This first VSM 4.0 applicationis namedthe “current VSM 4.0”, documentingthe status of each phase of the processtowardsthe smart operation and smart factory. Besidesthese I4.0 measures of smart operations and smartfactory,the classical VSM measures are also produced. Atthe sametime, someimprovement opportunities were registeredtowardsthe smart factory and smart operations, so that these improvements may help the company to attainthe strategy registeredinthe project charterform. A new VSM 4.0isthen e lab-orated, named “VSM 4.0 future state”, with estimated new values ofthe classical VSM items and smart operations and/or smartfactory new estimated values.
3.1.4 Step 4 – Evaluation ofthe conceptualframework.
In this regard, we will use the approach of Technology Acceptance Model (TAM)in orderto assessthe usefulness and ease of use ofthe proposed frame-work. The decision behind this validation approach is based on the methodological principles ofthe Information System theory called “Technology Acceptance Model”. Thistheory proposesthat when users are presented with a newtechnology,two major factorsinfluence whetherthey will useit or not(Davis, 1985): 1) Perceived usefulness (PU): statesthe degreeto which users believethat using atechnology would enhance theirjob performance; 2) Perceived ease-of-use (PEOU):referstothe degreeto which users believethat using a system would be free from effort. Table 5 presentsthe per-ceived usefulness and perceived ease-of-use evaluation format employedlaterinthis work.Inthis case, when possibleimprovementsissues have beenidentified, weind i-catethose considered priority. This qualitative evaluationis discussedin Section 7.1.2 (de Santa Eulalia, 2011).
Table 5: TAM criteria Tool aspect Perceived
use-fulness Improvementcomments Perceease-ofived-use Improvementcomments Impact on I4.0 strategy
deployment
Impact of data col lec-tion and analysis Impactin company per-formance
Impact in unders tand-ing I4.0 concepts
3.1.5 Step 5 –Improvethe conceptualframework
This step was not performedinthis research, duetotime constraints inthis Master’s degree, but future directions are proposed atthe end ofthis document.
4 Systemat
ic
L
iterature
Rev
iew
Resu
lts
Subsection 4.1 presents a bibliometric analysis, and Subsection 4.2 provides the content analysis of all articles.
4.1 Bibliometric results
The bibliometric analysis was based upon five criteria: year of publication, journal or conference, authors, country of the principal author, and research design. Figure 3 shows the articles by publication year. Concerning I4.0 maturity models (I4.0MM)the figure showsthe first year as 2014, which meansthatthis subjectis quite new, sincetheterm I4.0 was firstlaunchedin 2011. Theliteratureis stillinitsinfancy, with atotal of 6 articles until 2017. Regardingtheindustry 4.0 &lean manufacturing integration(I4.0 & LM),the first article was publishedin 2015, andtheinterestforthis subject has beenincreasing sincethen.
Figure 3 - Articles by publication years
Figure 4 presentsthe articles per periodictype. Regarding I4.0 MM,thereis a predominance of white papers (4 out of 6), which denotesthatthis subjectis still mostly covered by the grey literature rather thanthe scientific one. Concerning I4.0 & LM, thereis a slight predominance ofthe conference(6 out of 11) overjournals.
1 3 1 1 6 0 1 3 7 11 0 2 4 6 8 10 12 2014 2015 2016 2017 Total Nu mb er of art ic le s
Figure 4 - Articles by periodictype
Figure 5illustratesthe number of articles bythe country ofthe principal au-thor, with Germany predominating, for a total of 3 out of 6 articles concerning I4.0 MM, and atotal of 4 out of 11 concerning I4.0 & LM. It can also be noticedthatitis in Europe wherethese articles are mostly written,including other countrieslike Italy, Croatia and Poland. Other continents have startedto publish articles on this subject, such as America (USA, Brazil) and Asia (India and Turkey).
Figure 5 - Articles by country ofthe principal author
Figure 6 below showsthe number of articles by research methodology. Itis not surprising thatthe research strategy most used isthe qualitative one, based on sem i-structuredinterviews, becausetheimplementation data forthis subjectis scarce.
1 1 4 6 6 5 0 11 0 2 4 6 8 10 12
Conference Journal White paper Total
Nu mb er of art ic le s
Periodictype I4.0MM I4.0 & LM
3 0 0 0 0 1 1 0 1 6 4 2 1 1 1 0 1 1 0 11 0 2 4 6 8 10 12 German y Italy Poland Croati a
Brazil Indi a
Austri a
USA Turkey Total
Nu mb er of art ic le s Country I4.0MM I4.0 & LM
Figure 6 - Articles by research strategy 4.2 Content results
Table 6 showsthe content analysis regarding I4.0 MM. Regardingthetype of industry, all models were made for the manufacturing industry, except Industry 4.0/Digital Operations Self-Assessment - PWC model, which can be appliedin manu-facturing and otherindustries,like retail & consumer, financial services, etc. Concern-ingthe employed dimensions,the model withthe highest quantityis “A maturity model for assessing Industry 4.0”, with 8 dimensions, followed by the IMPULS and PWC model, with 6 dimensions each. Theless detailed modelregarding dimensions number is the I4.0 Reifegrad-model, with 3 dimensions. As for integration with LM, no ma-turity model presentedthis feature. Dataabouttheimplementation ofthese models are rare, butin the year 2015 the IMPULS provided some data about 289 companies in Germany. Thistoolisin evolution andthere should be more data nowadays concerning the number of companies, countries,etc. (Lichtblau et al., 2015).
3 1 1 1 6 5 1 3 2 11 0 2 4 6 8 10 12
Qualitative Quantitative Mix not available Total
Nu mb er of art ic le s
Research strategy
Table 6: I4.0MM content Analysis
Model IndusType otryf Ddescrimensipionstion number, Inwithtegrat LM Useion Data
IMPULS – Industrie 4.0 Readiness.
(Lichtblauet al.,2015) Manufacturing 6
Smart factory
no
289 in-dustries in Ger-many Smart operations
Smart products Data-driven services Employee
Strategy
Industry 4.0/ Digital Op-erations Self-Assessment -PWC model
(Pricewaterhouse Coopers ,2015)
Several 6
Business Models, Product & Service Portfolio
no avaNoilabtle Market & Customer Access
Value Chains & Processes IT Architecture
Compliance, Legal, Risk, Security & Tax
Organization & Culture
A maturity modelfor assessing I4.0
(Schumacher et al.,2016) Manufacturing 9
Strategy No. 1 indus-tryin Austria Leadership Customers Products Operations Culture People Governance Technology I 4.0 Reifegradmodell
FH – (Oberösterreich,
2015) Manufacturing 3
Data
no avaNoilabtle Intelligence
Digital
Transformation
I4.0-MM
(Gökalp et al.,2017) Manufacturing 5
Asset Management
no avaNoilabtle Data governance
Application management Processtransformation Organization alignment
Figure 7 organizes all articles by I4.0 & LM integration type. Information aboutthis subjectis still scarce, duetothelow number of articles foundintheliterature review (11). There were 9 articles out of 11that considered I4.0 as an enablerto LM, where I4.0 dimensions/technologies can remove some barriers related to traditional leanimplementations,like high demand volatility, high product variety and reducedlot sizes. Also, we foundthe LMtool as an enablertoI4.0, with 1 article out of 11, where an adapted LMtoollike Value Stream Map 4.0 (VSM 4.0)identifies digitalization op-portunitiesin a process, and also analyzes waste reductioninthe data/information flow of a process. Finally, in 1 article out of 11, both I4.0 and LM are complementaryto each other.
Figure 7 - Articles by I4.0 & LMintegrationtype
Table 7 reviewsthe I4.0 & LMintegration. Only 1 out of 11 articles presents LMtools adaptedto I4.0 (column LMtools adaptedtoI4.0). The columnI4.0 & LM integrationlevel describesthat 6 out of 11 articles give clues abouttheintegrationlevel, i.e.I4.0 dimensionslinkedto LM principles.
9 1 1 11 0 2 4 6 8 10 12 I4.0 as enablerto
LM enabLMtoolerl asto I4 an.0 compLM &lemen I4.0 ataryre Total
Nu mb er of art ic le s
Table 7: I4.0 & LMintegration summary #
Authors I4.0grat &ion LM typeinte- I4.0 & LMintegration details adapLMtedtoo tols I4.0
1 Dombrowski
et al. (2017) I4.0 as enabLM lerto
I4.0technologies,like cloud compu -ting,improve LM principle(waste avoidance).I4.0 process digitaliza -tion improves all LM principles, e s-peciallythe standardization.
n/a
4 Jayaram et al. (2017)
LM & I4.0 are complementary to
each other
LMandI4.0arecomplementary to
each other n/a
5 Kolberg et
al. (2015) I4.0 as enabLM lerto
I4.0 dimensions (smart operator, product, machine, planner)improves LM Principle (Justin Time,jidoka),
methods (Kanban, Andon)
n/a
7 Meudt et al.
(2017) LM as enabI4.0lerto n/a VSM 4.0
8 Lödding et
al. (2017) I4.0 as enabLM lerto
I4.0 dimensions (smart operator, smart product, smart machine, smart planner, smart workstation) im -proves LM Principles (JIT, total quality management, Total produc -tive maintenance, Humanresources management)
n/a
9 Mrugalska et
al. (2017) I4.0 as enabLM lerto
I4.0 Dimensions (smart products, smart machines, augmented opera -tor) improve LM principles (JIT, Poka-yoke,single minuteexchange die, continuous improvement, jidoka)
n/a
12 Rauch et al.
(2017) I4.0 as enabLM lerto
I4.0technologies,like cloud compu -ting, digitalization influence LM principles such as JIT
14 Sanders et
al. (2016) I4.0 as enabLM lerto
I4.0technologies as a solution to
problems at LM dimensions n/a
16 Tortorella et
al. (2017) I4.0 as enabLM lerto
Association between LMimplemen -tationlevel andI4.0technological
level on company´s performance n/a 17 Veza et al.
(2016) I4.0 as enabLM lerto
Correlation between companies’ performance, I4.0 and LMimple
-mentation n/a
18 Wagner et
4.3 Literature review discussion I4.0 as an enablerto LM
LMisthe foundation for I4.0(Dombrowski et al., 2017). To analyzethe in-terdependencies and correlations between Lean Production Systems and I4.0, 260 us -age cases at companies were analyzed, concerningthe application of I4.0technologies, process related characteristics of I4.0 and principles of Lean Production. I4.0techno l-ogies were defined as: big data, radio frequencyidentification (RFID), cloud compu-ting, augmented and virtual reality, sensor/actuator, real-time data, automated guided vehicles (AGV), consumer electronics. The process-related characteristics of I4.0 were described as: horizontal and verticalintegration, real-time data,transparency, flexibi l-ity, digitalization, consistency of information, monitoring, visualization, traceability and self-optimization. LM principles were defined as: standardization, zero defects, continuous flow, pullflow, continuousimprovement, employee orientation and man-agement by objectives, visual management and avoidance of waste. I4.0 process- re-lated characteristics were described as: horizontalintegration, verticalintegration, rea l-time data,transparency,flexibility, digitalization, consistency ofinformation, moni tor-ing, visualization,traceability and self-optimization.
Regardingthelink between I4.0technologies and LM principles,the findings were: 84 out of 260 companies that applied the LM principle of avoidance of waste indicated the usage of cloud computing as I4.0technology, so cloud computing was the most popularI4.0technology forthis LMprinciple. The zero defect LM principle indicated big data asthe most used I4.0technology, with 37 cases out of 152. Concern-ingthelink between I4.0 process-related characteristics and LM principles,the main results were:in 89 out of 499 casesthat appliedthe LM Standardization principle,the most-used I4.0 process was digitalization (Dombrowski et al., 2017).
Kolberg et al.(2015) and Mora et al. (2017) stated some LMlimitations, such as difficulties to level capacity utilization, due to strong variation in market demand. I4.0technologies, such as realtime-data collection and analysis, and autonomous pro-cesses, can mitigatethis LMlimitation,levelling capacity utilization automatically. As LM startedinthe 1950s,it does nottakeinto accountthe use of moderntechnologies, such as the I4.0 ones. In ordertoimprove LM principles and methods, Kolberget al. (2015), Moraet al.(2017) and Mrugalska et al.(2017) statedthat someI4.0 dimensions can beimplemented, such as: smart operator, smart product, smart machine, smart p lan-ner (Kolberg, Zühlke, Zuehlke & Zühlke, 2015). For example,the I4.0 smart operator can empowerthe LM methodslike Kanban and Andon. Kanbanis a method oflabelling small production lots to get better control of raw materials, purchased parts, work-in -progress as well as ofthe rate,total volume andtiming of production (Gravel & Price, 1988).
Smart operator helpstoimplementthe Kanban method, because operators can getinformation abouttheremaining production cycletime via augmented reality. An-donisa visual managementtoolthatshowsthestatus of operationsinanareaand signalizesthe occurrence of abnormalities (Kemmer, 2016). Withinthe Andon method, by whichemployeesshould be notifiedassoonas possibleincase ofafailure,the smart operator could reduce time between failure occurrence and failure notification. Equipped with smart watches, employees receive error messages and error locations close to real time. In comparison to widespread signal lamps, recognizing failures no longer depends onthelocation of employees.
Smart products can enable Kanban method because they contain Kanban in-formationto control production processes. Smart machines have standardized physical andITinterfaces, suitablefor sending andreceiving Kanbaninformation.It can support Andon by sending failures directlyto smart operators andinforming other systems for corrective action. Withthe smart planner,traditional Kanban systems withfixed
amount of Kanban, fixed cycletimes and fixed roundtrips fortransporting goodsturn into dynamic productions automatically adoptingto current production programs.
Rauch et al. (2017) presented research onthe I4.0influencein LM principles ina product developmentenvironment. Thefindings were: 55% ofrespondents de-clared that cloud computing, one of I4.0 technologies, influences the LM principle of interdisciplinary product development processes. Digitalization,like digital wo rk-flows, influences the LM JIT principle, was the response of 45% of the respondents (Rauch, Dallasega & Matt, 2016).
Sanders et al. (2016) analyzed how I4.0technologies and processes can con-tributeto LM dimensions,including supplierfeedback, JIT delivery by suppliers, sup-plier development, customerinvolvement, pull production, continuousflow, setuptime reduction, total productive maintenance, statistical process control and employee in-volvement.
One paper presented the correlation between companies in Brazil, regarding thefields of operational performanceimprovement,I4.0 technologylevels,and LM implementationlevels. The results have shownthat only for companies with high op-erational performance improvement, there was a significant association with LM and I4.0(Tortorella & Fettermann, 2017).
There has beenresearch about LM and I4.0implementationin Croatian com-panies. It was found that 75% of the companies do not apply LM, and that fact was detected as a main obstacle for companies to move towards I4.0, because LM imp le-mentation creates one ofthefoundationforI4.0implementation,reducing wastein processes, performing standardized work, visualization of performanceindicators, etc. (Veza, Mladineo & Gjeldum, 2016).
Implementation ofI4.0 providestheinsertion of newtechnologiesintothe current LM processes as well asthe change of business processes. Theinfluence ofI4.0 technologies on LM systems was researched in 24 companies in Germany. I4.0 tech-nologies were clusteredintothree Cyber Physical Productions Systems(CPPS), asfo l-lows: data acquisition and data processing, machineto machine communication(M2M) and human-machineinteraction(HMI). Data acquisition and data processing use tech-nologies such as sensors and actuators, cloud computing, big data, analytics, so that these hardwareandsensorscancommunicateandinteract withthe physical world. M2M containsthetechnologies of vertical and horizontalintegration. Verticalin tegra-tion connects machinesand data on different levels. Horizontal integration connects machines and data on the same level. Human-machine interaction (HMI) consists of information sharing and collaboration between machines and employees, viatechno l-ogieslike virtual reality and augmented reality (Wagner, Herrmann & Thiede, 2017).
LMtools as an enablerto I4.0
The Value Stream Map (VSM) is a method that allows the analysis of value addedin a process chain, sothat waste elimination/reduction opportunities can be iden-tified and addressed. An adapted version of VSM, the VSM 4.0, can identify digital improvement opportunities. The analysisis also extendedtoidentify wastesin data and information, which differsfromtheclassical VSM, which generally onlyidentifies wastes in the physical activities of a process. It also analyzes the collection point of data/key performance indicators (KPI), its storage media and where they are used. It providesthe following metrics: data availability(DA), data usage(DU) and digital isa-tion rate (DR). DA stands for the % of the planned data points that are actually co l-lected. DU means the % of the planned data points that are used for continuous im-provement or decision-making. DR means the % of the data collected that is digital (Meudt et al., 2017).
4.4 Research gapsidentifiedintheliterature
The integration between I4.0 and LM was presented only at the operational side,there was notype ofintegration regardingthe strategic side, showing howto de-ploy I4.0 strategyintothe shop floor withthe help of LM principles ortools, whichis aresearch gaptoday. Another gapisthat only 1 out of 11 articles demonstrated how LM adaptedtools can helptoidentify and quantifytheimplementation potential of I4.0 technologies,interms of waste reduction, qualityimprovement, etc.
5 THEORETICAL
FRAMEWORK:
LSS
ADAPTED
TOOLS
The adapted model of Erol et al. (2016) will serve as a comprehensive f rame-work forthe strategy ofthe I4.0. The adaptationthat was made wasto relatetools for each phase ofthe model, accordingto Figure 8, becausethe original framework did not havethe meansto make concrete andtangiblethe principles ofI4.0 atthe operational level, sothetools makethe bridge betweenthe strategic and operationallevel.
Thefirst phase, envision, stands forthe companies common understanding of its current maturitylevelregardingI4.0. Thetool used forthis phaseistheIMPULS method andits questions are shownin Annex 1. Itis appropriate, becauseit coversthe different dimensions ofindustry 4.0 proposed by Schumacher, Erol & Sihn (2016) and represents several principles of I4.0 proposed by Hermann, Pentek & Otto (2016). Ta-ble 8 presents a match ofthese principles andthe dimensions ofthe IMPULSto d iag-noseits current stateinthe company.
Table 8: IMPULS dimensions and I4.0 principles IMPULS
dimensions I4.0 principles
Smart products Informationtransparency decentralized decisions Smart operations Interconnection, decentralized decisions
Smart factory Informationtransparency Data-driven services Not applicable
Employees Notapplicable Strategy & organization Not applicable
The second phase, enable, meansthe visionforthe I4.0 strategy forthelong term andits road mapping. Thetool adopted forthis phaseisthe project charter, which is described later in Section 5.2. Finally, the third phase, enact, consists of the I4.0 strategytransformationin projects. Thetool forthis phase arethe LSS-adaptedtools tothe I4.0 context,that are presentedlaterin Section 5.3
Figure 8 -I4.0 strategy framework. Adapted from Erol et al. (2016) 5.1 I4.0 Envision – IMPULS diagnostic audit
IMPULS defines sixlevels of maturity oftheI4.0(0-newincoming 1-beg in-ner 2-intermediate; 3-experiment; 4-expert; 5 -top performance). The IMPULS also presentsthe main obstaclesto achieve a higherlevel of maturity. TheIMPULS ques -tionnaireis shownin Annex 1.
5.2 I4.0 Enable – Project Charter
The project charterformis necessaryto documentthe companyI4.0 strategy commitment and main projects, as well asits main deadlines. This document serves as a guidetoimplement I4.0 projects, becauseit presentsthe I4.0 dimensionsthat should be prioritized. Theform hastwo main parts accordingto Figure 9: Part 1 stands forthe I4.0 current state, wherethe current company I4.0 maturitylevelis registered, as well asthe I4.0 dimensions maturitylevel. This maturitylevelsinformation comes fromthe IMPULS report, afterthe company has respondedtothe IMPULS questionnaire. The second parts standforthe I4.0 desired future state,interms of maturitylevel, andthe
• Impuls diagnostic audit
• Project charter
• LSS adaptedtoolsto I4.0
Tools I4.0 strategy
Envision • Common understanding of I4.0
Enable • Roadmapping of I4.0
I4.0 dimensionsthat should be prioritized. The action planto attainthese future I4.0 maturitylevelsis also documented. The other supportitems ofinformation are Current I4.0 projects/initiatives, andthe VSM as-is findings and VSMfuture state (Pyzdek & Keller, 2010).
Figure 9 -I4.0 project charter, adapted from Pyzdek et al. (2010) 5.3 Enact - LSS adaptedtools to I4.0
5.3.1 Classic VSM
Value Stream Mapping (VSM) aims to allow systematic identification of losses and wasteinthe production process. It allows representingthe actionsthat create value andthosethat do not create valueinthe process oftransformation of a product from aninitial stateto afinal state. It also allowsthe detection of potential improve-ments (Meudt et al., 2017). Value stream mapping promotesthe visualization of station cycletimes,inventory buffers atintermediate stations, manpower usage, value added
percentage,availabilityrateandcycle time.It recordstheentiretransformation ofa productionline,from raw materialsto finished goods. The VSMis normally recorded using an EA3 (11x17inch) size paper. There areiconsthat representthe customer and its shippingfrequency,the supplier anditsreceivingfrequency,the production control anditslink with suppliers and customers (Seth & Gupta, 2005).
VSM is based on 5 phases: 1) selection of a product family; 2) current state mapping; 3) future state mapping; 4) defining a working plan; 5) achieving a working plan. Some guidelines are necessaryforthe definition ofthefuture state map,including the one saying that the production rate should follow customer demand (“takt time”), borrowedfromthe German Word “Takzzeit”, which means clockinterval. Continuous flow should be established whenever possible; wherethe continuousflowis not poss i-ble, employthe pull system (wherethe productionis pulled accordingtothe customer demand), which differsfromthe non-lean method(where productionis pushed accord-ingtoeachstationcapacity, generating unnecessaryinventory). Only one process, calledthe pacemaker or bottleneck process, should commandthe production of diffe r-ent parts, andit will setthe paceforthe whole value stream.
The VSM hasthefollowing advantages: The analysis of as-is stateisbased on the collect and analysis of numerical data andit uses a visualinterface whereitis easy toseetherelationship between materialandinformationflows. Theanalysis ofthe whole value chain of a productfamily allows seeingtheinefficiencies. The delivery of a standardlanguagefortheteam andthe unification oflean concepts andtechniquesin a unique body. The VSMcan be the beginning ofastrategic planforimprovement (Serrano Lasa, Ochoa Laburu & de Castro Vila, 2008)
5.3.2 VSM adaptedtoI4.0 – VSM 4.0
Accordingto Figure 10,the adapted VSM 4.0 has a classical part, anindustry 4.0 (smart operations, smart factory), and a businessintelligence (BI) part, shownin Figure 11.
The adapted VSM 4.0 is shown in Figure 10. The following sections were addedintothe classical VSM: smart operations, smart factory. Smart factory stands forthe following sub dimensions: Equipment Infrastructure, Data Collection, Data us -age. They were addedto reflectthe deployment ofI4.0 strategy, becausethese d imen-sions are found on IMPULS method, whichisthe first steptowards anI4.0 strategy, accordingto Figure 8. Oncethese I4.0 sections are addedinthe classical VSM, itis possibleto checkthe strategy deployment onthe shop floor. Table 9 showsthe eva lu-ation grid for each one ofthese sections.
Data collection and data usage details
Data collection and usage are shownin Figure 11,following BI process con-cepts. This section was adapted from Meudt et al.(2017) wherethefollowing sections were added: data quality, cost, visualization, analysis and decision. The fields are ex -plainedin Table 10. Data quality has a specific grid, detailedin Table 11,it assesses some factorsthat mayinfluence data quality, such as:type of storage(paper or digital), type of recording for digital storage (semi-automated or automated), presence of qua l-ity verification for non automatic recording.
Figure 10 - VSM 4.0 – Adaptedfrom Meudt et al.(2017)
Value Stream Mapping 4.0
Delivery schedule type Production schedule frequency
TASK 1: TASK n:
Receiving frequency # of persons: # of persons: Shipping Frequency
Shifts: Shifts:
Cycle Time Cycle Time
Value added% Value
added% availability
rate: availability rate:
Lot size Lot size
Equipment: Data collection: Equipment: Data collection: Inventory Level
Data usage: Data usage:
Autonomous
Process Automation Autonomous Process Automation Information Sharing: Information Sharing:
Lead time cycle time Lead time cycle time Lead time VSM CLASSICAL SMART FACTORY SMART OPERAT. Production control Production supervisor SMART FACTORY SMART OPERAT. VSM CLASSICAL Customer Supplier
Table 9 - IMPULS evaluation by dimension, adapted from Lichtblau et al. (2015) IMPULS
DIMENSION DIMENSION SUB- VSM Action
SMART
FACTORY infrasEquipmentructuret
0-Machine and systeminfrastructure cannot be controlledthrough IT, no M2M
1- Some machines can be controlledthrough IT, areinteroperable, or have M2M capability
2- Machine and systeminfrastructure can be controlledto some ex-tentthrough IT,isinteroperable orintegrated
3-Machine and systeminfrastructure can be controlledthrough IT andis partiallyintegrated
4- Machinery can be controlled completelythrough IT,is partially in-tegrated(M2M) orinteroperable
5-Machines and systems can be controlled almost completelythrough IT and arefullyintegrated (M2M)
SMART
FACTORY Data collection
0 - no datais collected 1- no datais collected
2- Datais collected but forthe most part manually 3- The relevant datais collected digitallyin certain areas 4- Comprehensive digital data collectionin multiple areas 5- Comprehensive, automated, digital data collectionin all areas
SMART
FACTORY Data Usage
0 - no data available for further use 1 - no data available for further use 2- Datais usedfor a few select purposes (greatertransparency, etc.)
3- Some data usedto optimize processes (predictive analysis), 4- Data usedin several areas for optimization
5- Data usedin all areas for process optimization
SMART
OPERATIONS Auprocessestonomous
0-Autonomously guided workpieces notin use 1-Autonomously guided workpieces notin use 2-Autonomously guided workpieces notin use 3-Autonomously guided workpieces notin use 4-Experimentsintest and pilot phase
5-Usein selected areas or even cross-enterprise
SMART
OPERATIONS Informasharingtion
0-No system-integratedinformation sharing
1-Beginnings ofin-company, system-integratedinformation sharing 2-In-companyinformation sharing partially system-integrated 3-Somein-company and beginnings of external system-integrated in-formation
4-Predominantlyin-company and partially external system-integrated information
5-Comprehensivein-company and partially external system-in te-gratedinformation sharing
