Exploring underlying mechanisms influencing Smart & Lean manufacturing performance: a multiple case-study

(1)

Exploring underlying mechanisms

influencing Smart & Lean

manufacturing performance: a

multiple case-study

June 24, 2019 Word count: 12.102 Tim de Roos S2578530

Master of Science, Technology and Operations Management Master Thesis TOM

(2)

Abstract

The everlasting desire to seek maximal operational performance is entering a new era by combining lean and smart manufacturing. Mass-customization is challenging the flexibility of lean manufacturing, while smart manufacturing is promising improved flexibility. This multiple case-study explores and describes underlying interactions mechanisms of lean and smart influencing operational performance in a high variety and low volume organizational context. Case observations and interviews revealed several examples of interactions mechanism of lean and smart in the current and future state. The interaction mechanisms of smart and lean manufacturing are found to be able to support or complement each other and thus can improve operational performance. Moreover, smart can even become a driver for lean manufacturing. However, smart manufacturing should be built on the foundations of lean. Smart manufacturing is mainly speeding up the flow of information, which enables an increase in flow of materials and flexibility. Currently, vertical integration and standardization through smart technologies improves flows of material and information. In the near future, horizontal integrations will create a stronger customer-supplier link throughout the entire supply chain enabled by interconnected smart technologies. Additionally, the fact that the critical success factors of lean and smart are matching is implying that a company successful in lean, has a good change to be successful in smart as well. Management, resources, and knowledge do positively influence the level of interactions between lean and smart manufacturing.

Keywords: smart manufacturing; lean manufacturing; operational performance; high variety

(3)

4. RESULTS ... 19 4.1 CURRENT STATE... 19 4.1.1 Company A ... 19 4.1.2 Company B ... 21 4.1.3 Company C ... 22 4.1.4 Company D ... 24 4.2 SUCCESS FACTORS ... 25 4.2.1 Management ... 25 4.2.2 Resources ... 27 4.2.3 Knowledge ... 29 4.3 FUTURE STATE ... 30 4.3.1 Company A ... 30 4.3.2 Company B ... 31 4.3.3 Company C ... 32 4.3.4 Company D ... 33 5. DISCUSSION ... 34 5.1 MAIN FINDINGS ... 34 5.2 PRACTICAL IMPLICATIONS ... 36 6. CONCLUSION ... 37

6.1 ANSWERING THE RESEARCH QUESTION... 37

6.2 LIMITATIONS AND FUTURE RESEARCH ... 37

REFERENCES ... 39

APPENDIX A–INTERVIEW PROTOCOL ... 44

APPENDIX B–CODING SCHEME ... 46

(4)

Preface

This master thesis marks the end of my academic journey by completing the Master of Science in Technology & Operations Management. Five years ago, starting with a Bachelor of Science in Biology & Medical Laboratory Sciences, I never thought that I would complete a master at a university level in a business-related educational field. This thesis, strongly rooted in the cores of Operations Management has brought me joy and a solid understanding of what is to come in our field of research.

Nonetheless, this would not have happened without the help and support of some people. First, I would like to thank all interviewees of the case companies to dedicate their precious time for this research. The interviews were highly informative mainly due to their contributions.

Second, my sincerest gratitude to Mr. Knol and Ms. Nijhof of the Hogeschool Arnhem & Nijmegen and Mr. Nanninga from the Hanze Hogeschool. Not only did they provide the case companies, but they also were my partners during the case studies and interviews.

Third, I would like to thank my family and friends. In particular my girlfriend, Marije, who has been greatly supporting me and made it possible to dedicate my time to successfully complete this journey.

Finally, I have to thank my supervisor dr. J.A.C. Bokhorst for his highly valuable feedback, guidance and time. I can honestly say that this thesis is at this level because of you. Sincerely,

(5)

1. Introduction

Over the past decades, every company seeks to find maximal operational performance. Among lean manufacturing (LM), many philosophies and principles, like Six Sigma or Total Quality Management have emerged in order to increase operational performance. Recent developments in technology gave rise to what is considered the fourth industrial revolution: Industry 4.0, often referred to as Smart manufacturing (SM) (Kang et al., 2016). Smart manufacturing technologies like Cyber-physical systems (CPS), Internet-of-Things (IoT), Cloud Computing, and Big Data are believed to make mass-customization possible. However, it remains unclear whether lean principles and smart manufacturing technologies can actually strengthen each other (Kolberg and Zühlke, 2015; Wagner, Herrmann and Thiede, 2017). Organizations that have been investing heavily in lean do not want to risk decreased operational performance by introducing smart technologies. This study therefore aims to explore underlying mechanisms between smart manufacturing and lean manufacturing influencing operational performance.

Extensive research has been conducted on how to implement and utilize lean manufacturing to its fullest potential (Shah and Ward, 2007; Sundar, Balaji and Satheesh Kumar, 2014). On the contrary, the literature on smart manufacturing is immature but rapidly developing (Buer, Strandhagen and Chan, 2018). More recently, academics are diving into the field of using smart technologies as a complementing instrument in lean manufacturing (Kolberg and Zühlke, 2015; Wagner, Herrmann and Thiede, 2017; Tortorella, Giglio and Dun, 2018). However, contrasting findings are available in the current literature frontier on synergies between both philosophies (Hof, 2017; Tortorella, Giglio and Dun, 2018). While Hof (2017) found that utilizing smart and lean together does increase operational performance, Tortorella

et al. (2018) found that it depends on the sort of interaction. Jointly exploiting lean and smart

(6)

Due to the immaturity of smart manufacturing literature, it has to be investigated how smart manufacturing can complement lean principles in order to improve performance. Synergies have been identified but, more importantly, the underlying mechanisms need to be revealed (Hof, 2017; Kühl et al., 2018). Moreover, it is unknown how and what kind of smart manufacturing technologies can be applied in a lean production environment (Sanders, K. Subramanian, et al., 2017) . Hence, in order to take the next step in future manufacturing, it is crucial to explore how lean and smart manufacturing interact. However, in exploring how LM and SM should interact, we cannot deny the success factors or barriers impacting what kind of interactions may exist. Therefore, in addition, the organizational context (i.e. management, resources and knowledge) has to be considered in an environment where lean principles are already in place while smart manufacturing is to be introduced as a complementary factor.

In order to fill this gap in the literature, a multiple case study will be used to explore interaction mechanisms influencing lean and smart manufacturing performance. By using these findings, practical and managerial implications can be derived on how to utilize smart and lean manufacturing to increase operational performance. Additionally, the success factors of the organizational context impacting the interactions mechanisms between lean and smart will be targeted. Four companies in a high variety and low volume process type will be studied as they may benefit most from the big promise of smart manufacturing: improved flexibility (Moeuf et

al., 2017; Zhong et al., 2017; Buer, Strandhagen and Chan, 2018). The main research question is how are underlying mechanisms of smart and lean influencing operational performance.

(7)

2. Theoretical Background

This section provides an overview of the current literature in relation to the research question. First, a brief history of lean manufacturing is given, accompanied by the definition, its main principles and implementation drivers and barriers. Second, smart manufacturing will be discussed with regard to the history resulting in its current state, the varying definitions, available technologies and implementation barriers. Third, an extensive overview is given on the known impacts of combining SM & LM on operational performance. Lastly, the conceptual model is developed, described, and linked to the research questions.

2.1 Lean manufacturing History

Originating from the Toyota Production System (TPS), lean thinking was developed on the Japanese shop floors. This approach focuses mainly on the elimination of waste in terms of non-value adding activities (e.g. transportation, inventory, motion, waiting, over-production, over-processing, defects). The term ‘lean production’ (i.e. lean manufacturing) was first mentioned in the book ‘The machine that changed the world’ (Womack, Jones and Roos, 1990). Although the underlying principles (e.g. just-in-time and Kanban pull production) were established prior to this work, the book acted as a starting point to shift the operational ideals on how to achieve high customer value (Hines, Holweg and Rich, 2004).

Definition

(8)

Principles & Practices

Four basic rules have emerged in lean manufacturing from the TPS: (1) work standardization; (2) uninterrupted work flows; (3) direct customer-supplier links; and (4) continuous improvement (Spear and Bowen, 1999; Shah and Ward, 2007). These basic rules subsequently translate into many manufacturing practices and tools (e.g. single minute exchange of dies and value stream mapping). They can be grouped into four different bundles that have a high inter-relation and consistency of practices (Shah and Ward, 2007): (1) Just-in-time (JIT): only the necessary quantity of products at the right time (Sugimori et al., 1977); (2) Total Quality Management (TQM): an approach that creates customer satisfaction by improving quality through continuous improvement of all processes (Flynn, Schroeder and Sakakibara, 1994); (3) Total Productive Maintenance (TPM): “an approach to maintenance that optimizes equipment effectiveness” (Ahuja and Khamba, 2008); and (4) Human Resource Management (HRM).

Lean manufacturing is mostly found to increase operational performance in terms of: flow, flexibility, and quality (Shah and Ward, 2003; Taj and Morosan, 2011). However, the basic rules and practice bundles of the TPS which enable this performance increase were developed in a make-to-stock (MTS) process type environment (White and Prybutok, 2001) with repetitive processes. Contrastingly, in a make-to-order (MTO) job shop that is typified by a high variety of products produced in low volume, it is difficult to standardize non-repetitive work and to create a flow by pull production. Hence, lean principles have to be altered to fit the needs in MTO production. For example, the work-in-process regulating lean production tools like Kanban have been adjusted for MTO in Quick Response Manufacturing (QRM). A Kanban alternative, the paired-cell overlapping loops of cards with authorization (POLCA) system, is constraining the number of orders in the system by only releasing cards when an order has left the shop floor and capacity is available (Suri, 2010). Additionally, Slomp, Bokhorst and Germs (2009) found in a case study that slightly adjusted lean principles like combining CONWIP, FIFO and takt time can still significantly reduce cycle times in a low volume high variety environment.

Implementation

(9)

are proposed as a one-way approach, lacking a contingency sense to be able to adapt it to a specific case. Nevertheless, the five phases in the dynamic model of Anvari et al. (Anvari et

al., 2011) can be used to assess the lean maturity of any organization (Maasouman and Demirli,

2016): (0) Initial investigation: there has to be a need for improvement or crisis (e.g. in profit), management has to be committed, and knowledge about lean is required to be able to start integrating lean; (1) Preparation: assessing the level of knowledge among all employees (e.g. experts and shop floor) and the strategic choices of an organization; (2) Pilot: draw both a current and future value stream map for one stream and identify opportunities for waste reduction. Here, the lean toolbox is used to create flow, stability, flexibility, and pull; (3) Expand: stage 2 is repeated for all organizational processes, including the office and supply chain; (4) Perfection: measuring and monitoring performance and establishing an organization-wide lean culture.

Figure 1, Lean transformation framework (Anvari et al., 2011)

(10)

Among the ‘hard’ lean practices (e.g. Kanban and statistical process control), soft practices (e.g. culture and leadership) are required to successfully implement lean (Bortolotti, Boscari and Danese, 2015). In other words, a lean-thinking organizational culture, which includes both management commitment, employee involvement, and empowerment is key to success and thus improves performance. In order to successfully create such a lean organization that is capable of executing both hard and soft lean practices, four implementational barriers have to be overcome: (1) management: which has to make a commitment to change and create organizational readiness; (2) resources: to provide financials, employees and time to be able to change; (3) knowledge: to develop skills and expertise in lean; (4) employee engagement: to develop lean thinking culture on the long-term and minimize conflicts (Almanei, Salonitis and Xu, 2017; Pearce, Pons and Neitzert, 2018). If these barriers are overcome, they become success factors for lean manufacturing implementation and thus enable an increase in performance. 2.2 Smart manufacturing

History

The potential of manufacturing enabled by smart technologies was first recognized in the form of ubiquitous computing (Zuehlke, 2010). Ubiquitous computing, which comprises intelligent devices applied in any setting, is considered as the key to solve the need for real-time data decision-making in manufacturing (Wang, Ong and Nee, 2018).

Definition

(11)

Technologies

Cutting-edge ICT technologies are presumed to deliver a competitive advantage and sustainable growth. Six main group technologies can been identified: (1) Cyber-physical systems: computational entities in intensive connection with the physical world; (2) Cloud computing; (3) Internet-of-Things (IoT); (4) Big Data analytics; (5) Simulation; and (6) Virtual Reality (Kang et al., 2016; Kusiak, 2017; Moeuf et al., 2017). A commonality is that sensors are crucial to provide these main group technologies with the correct data from the physical world (Kang

et al., 2016).

Most companies are found to limit themselves to cloud computing and internet-of-things as they are relatively inexpensive (Moeuf et al., 2017). However, as technologies continue to become cheaper over time, it is interesting to understand the mechanisms and potential of the other smart technologies. A strategic approach on how to implement this variety of technologies is required (Kang et al., 2016).

Implementation

The key ingredients to assess the readiness of an organization to implement SM are found to be: top management involvement and commitment, employee adaptability with SM, the extent of digitization of the supply chain, level of digitization of the organization and the readiness of the organizational strategy (Valaei and Rezaei, 2016). Additionally, factors influencing the success of introducing SM are: usability, selective provision of information, acceptance of users, consideration of ethical, legal and social impacts and profitability (Kühl et al., 2018).

Once an organization is ready to start implementing SM technologies, a pattern seems to emerge according to Frank et al. (2019). Three stages, in which the complexity of implementation increases, have been identified ranging from vertical integration in stage 1 to flexibilization in stage 3. The base technologies of cloud computing, IoT, big data, and analytics are input to the front-end technologies of smart manufacturing, smart working, smart products and a smart supply chain. Similar to the lean implementation framework of Anvari et al. (Anvari

(12)

Figure 2, Implementation patterns SM (Frank, Dalenogare and Ayala, 2019) 2.3 Combining lean and smart manufacturing

A question that is recently being raised in academic literature is whether or not lean and smart support each other (Mrugalska and Wyrwicka, 2017; Kühl et al., 2018). Several use-cases are showing that they can support each other (Dombrowski, Richter and Krenkel, 2017; Sanders, Subramanian, et al., 2017). For example, JIT and its supporting Kanban system is known to be digitalized (i.e. e-Kanban) and interacts with a CPS system to send order signals in real-time (Kolberg and Zühlke, 2015). This e-Kanban system acting in a CPS automatically detects empty bins, sends standardized signals to trigger machines or to reconfigure production lines. Furthermore, by using machine to machine communication, a gapless information flow is created to manage orders and materials. Sensors connected to raw materials (i.e. smart products) are detected in real-time to track material flow and stock levels. The big data from these sensors feed the CPS to autonomously align customer orders with supplier deliveries and schedules shop-floor operations in real-time (Netland, 2015; Mrugalska and Wyrwicka, 2017; Wagner, Herrmann and Thiede, 2017). Other major positive side effects of using an alike system is improved traceability, transparency and flexibility. Hence, finding the root cause of waste is easier and one-off customer specific products can potentially be produced in a profitable manner (Kagermann, Wahlster and Helbig, 2013).

(13)

smart technology has a positive effect on lean manufacturing. However, not every tool is considered to be as effective or supporting as the other, which was mapped in their impact matrix. For example, real-time data positively supports kaizen (i.e. continuous improvement), jidoka (i.e. automation), and TPM, but positive and negative findings exist on the effectiveness of big data on takt-time (Sanders, Subramanian, et al., 2017; Wagner, Herrmann and Thiede, 2017). An explanation for these contrasting findings of smart technologies frustrating the takt-time could be that the modularity and decentralized real-takt-time decision-making by a CPS might introduce extra manufacturing variability (e.g. schedule instability). Production schedules that are constantly changing in real-time on the basis of big data can negatively impact throughput times. For example, sudden schedule changes may introduce extra set-up times. Hence, the constant flow of materials is reduced. In other words, a standardized and constant flow of materials is still beneficial. On the other hand, Sanders et al. (2017) state that the lean tool takt-time may become obsolete with the introduction of SM as it does not allow for real-takt-time flexibility in scheduling and capacity. Changes can only be made after a cycle set by the takt-time ends. Thereby, it potentially blocks machines and services. Hence, one has to carefully consider what kind of SM technologies should be exploited in lean-thinking organizations and how the interactions would benefit the process.

Another challenge is to identify the best sequence to implement both management philosophies. First lean and then smart, vice versa or jointly together. Bortolotti & Romano (2012) state that it is beneficial to first map the process and identify waste and subsequently automate. Using this strategy, one will only automate value-adding operations. Thus, advocating for first lean and then smart (i.e. automation). This is further supported by Landscheidt & Kans (2016) and Tortorella et al. (2018), who state that badly structured manufacturing processes do not benefit from implementing cutting-edge technologies. In comparison, other academics think that industry 4.0 helps making their shop-floors lean (Sanders, Elangeswaran and Wulfsberg, 2016), implying that it could also work to first have smart manufacturing and then become lean. This makes sense, as otherwise the process will not benefit from the firstly implemented technologies. Smart manufacturing, which essentially is a form of automation, can magnify the shortcomings of processes (Bortolotti and Romano, 2012). Hence, implementing smart manufacturing could dictate to become lean in order to catch the full potential in efficiency gain through process automation.

(14)

resources in order to create an organization and culture that is both lean and smart minded. In turn, resources, which we here define as both financial and human resources (i.e. employees) are thus another critical factor to successfully implement both philosophies. Moreover, it is crucial to have a clear understanding of both philosophies. Lacking knowledge about either one or both of these philosophies will evidently lead to a lean and smart manufacturing mismatch. Hence, success factors can thus be a driver or barrier to successfully exploit lean and smart manufacturing.

The technologies of industry 4.0 can, as elaborated in the previous section, support and stabilize lean manufacturing and are mostly found to have a positive impact on LM (Mrugalska and Wyrwicka, 2017; Wagner, Herrmann and Thiede, 2017). However, a common framework linking smart and lean manufacturing is missing (Zuehlke, 2010; Kühl et al., 2018). Moreover, the impact is often discussed on a generic level (Prinz, Kreggenfeld and Kuhlenkötter, 2018) and lacking in-depth analysis of the underlying mechanisms (Kühl et al., 2018). “As Industry 4.0 aims at speeding up flows of information and Lean Manufacturing focuses on the elimination of waste to speed up physical flows, the synergy between the two methods should be considered to target operational excellence” (Moeuf et al., 2017). This is exactly what this research tries to contribute to academic knowledge: creating a better understanding of the underlying mechanisms influencing operational performance when smart & lean manufacturing are combined.

2.4 Conceptual model

(15)

The organizational context, consisting of the success factors management, resources, and knowledge has an impact on the interaction mechanisms. In this conceptual model it comprises of the success factors management, resources, and knowledge. As previously elaborated, these factors can affect implementation and thus impact the interaction mechanisms between smart and lean manufacturing. Direct relations of smart manufacturing and lean manufacturing on operational performance are out of scope. An illustration of the conceptual model is presented in figure 3.

Relating back to the developed conceptual model, we derive the following sub-questions in order to answer main research question: “How are underlying mechanisms of smart and lean

influencing operational performance?”

1. What smart and lean manufacturing aspects are being used and how do they interact? 2. What is the impact of the success factors in the organizational context (i.e. management,

resources, knowledge) on the interaction mechanisms of SM & LM?

(16)

3. Methodology

This section elaborates on the research set up and justifies the chosen methodology. First, we motivate the research design choice (i.e. case study). Second, an overview of the data collection is provided, accompanied by a description of how the data was collected. Third, we develop a methodology to analyze the collected data and draw conclusion accordingly.

3.1 Research design

Concluding from the theoretical background and the introduction, a solid understanding of what underlying mechanisms of lean and smart manufacturing influence operational performance is missing (Kühl et al., 2018; Prinz, Kreggenfeld and Kuhlenkötter, 2018). Furthermore, it is particularly important to understand how these mechanisms influence the performance. These open-ended questions are typically answered in the form of a case study (Karlsson, 2016). Although what questions could also be researched in a survey (Yin, 2003), it is in this context not deemed appropriate. We already concluded that synergies do exist (i.e. what), but we do not understand how they influence operational performance. Thus, identifying what underlying interaction mechanisms exist and how they actually increase operational performance would contribute to the body of knowledge. Hence, an explorative approach is required, which can be achieved by a descriptive and qualitative case-study (Dul and Hak, 2007).

3.2 Data collection

Case studies often rely on multiple sources of evidence (Yin, 2003). Therefore, in order to reach both internal and external validity, multiple cases are being analyzed. Four different cases have been observed. Each company produces different products in a high variety and low volume production environment. In this way, consistency in results can be achieved and validated. Additionally, as can be derived from the conceptual model, the organizational context is considered to influence interaction mechanisms.

Semi-structured interviews were used to explore the phenomenon at hand and to collect the required data. This primary data is derived from the interviewees and observations of the author. Furthermore, the interviews have been recorded.

(17)

understanding of the market demand, we asked what is required in terms of: speed of delivery, delivery reliability, products variety or complexity, quality and how this might change in the future. Second, we investigated how currently lean and smart manufacturing play a role in order to suit the previously identified market requirements. We considered what their impact is on operational performance (i.e. flow, flexibility, and quality) and where or how they interact. Third, the same questions were asked for the desired future state. What smart technologies will be required and how would this fit into lean processes in order to bring their specific operations to the next level? Fourth, considering the organizational context, we raised questions like how committed management is, if they have enough resources for integration projects regarding lean and smart, and whether or not there is sufficient knowledge of lean and smart manufacturing. Please refer to Appendix A for the detailed interview protocol.

3.3 Data Analysis

Data analysis started with a transcription of all interviews. Subsequently, the data has been analyzed and coded in Atlas.TI™ to create a better understanding of the results. Appendix B provides an overview of the used coding scheme. If any irregularity in the results existed, a check-up has been executed to guarantee a mutual understanding. Here, we do have to note that due to the explorative approach, an open-minded and wide perspective was required to qualitatively describe the observed interactions mechanisms. On the basis of the data coding and analysis, patterns or similarities emerged. In the end, this created a greater understanding of the underlying interactions mechanisms that influence operational performance.

3.4 Case selection and descriptions

(18)

Company A

The first case is a service provider in soil and water analytics. The company has a production facility that melts, cuts, bends, and assembles a high variety of products.

Company B

The second case is a metal production site, mainly performing cutting, bending and welding operations. The company’s main business is a high variety and low-volume environment for relatively small materials, although one-off projects are also being produced.

Company C

The third case is another metal production site. However, this company operates in the ship production industry. They cut and bend the parts of a ship and deliver it in subsequent order packages to the customer. Hence, the main operations are in high variety and of low volume. Additionally, a second and smaller part of the business is concerned with architectural structures, which typically are projects.

Company D

(19)

4. Results

This section will present the results from the multiple case studies. First, section 4.1 elaborates on the current lean & smart aspects of the individual cases and describes the influence of interaction mechanisms on operational performance. Second, section 4.2 is discussing the influence of the success factors on the interaction mechanisms between LM and SM. Third, in section 4.3, an elaboration of the desired future state of the companies is provided together with their interaction mechanisms.

4.1 Current state

In order to gain a better understanding of lean and smart maturity in the individual cases, a description of their lean and smart aspects is firstly given. These aspects are subsequently linked to the transformation frameworks of Anvari et al. (Anvari et al., 2011) and Frank et al. (2019) to assess the maturity in lean and smart manufacturing, respectively. Second, a description of the interactions mechanisms of these previously identified lean and smart aspects is provided and how these mechanisms influence performance in terms of flow, flexibility, and quality.

4.1.1 Company A

Aspects of Lean & Smart

(20)

With regard to smart manufacturing, there is currently little to find apart from: an ERP system with integrated shop floor control; a metal cutting machine that is connected to ERP and automatically loads part drawings; a warehouse picking system that guides the pickers on the most efficient routing; and a planning that indicates part availability. This immaturity of SM is mainly due to the fact that the management team is still searching for a common vision. Hence, company A is immature in smart manufacturing and scaled into stage 1 (vertical integration) of the Frank et al. (2019) framework.

Interaction mechanisms of lean & smart

Although we only observed a small number of smart tools in company A, there are some mechanisms in place. First, digital shop floor control systems vertically integrate the shop floor employees and quickly connect relevant information from all departments for each individual order. Hence, the flow of information is enhanced and standardized. Drawings can be visualized in 3D on the screen of the shop floor system and routings are digitally available. Thereby, a standardized information flow is created with the shop floor. Second, the automation of loading drawings is reducing information inertia by interconnecting the ERP system and the cutting machine and thus increasing speed. Third, the warehouse picking system calculates the best possible route to pick raw materials. Therefore, wastes like motion and transportation are minimized while performing these operations and thus flow of products is improved. Fourth, the planning system visualizes the status of raw materials requirement per order. Hence, it enables the production planner to identify problems fast and easy. Overall, these interaction mechanisms speed up the flow of information and increase vertical integration. Thereby, it enables an increase in the flow of materials.

(21)

4.1.2 Company B

The second case is currently working with three different QROCs (Quick Response Office Cells) following the QRM philosophy as form of lean management. Three customer clusters are being used to maximize customer value and speed. There are three work cells for each value stream. A cycle time reduction of 3 weeks was established by integrating this philosophy. This is mainly driven by their customers who demand faster and especially more reliable delivery of the products. Flow and structure have been created by standardizing operations on the shop floor. For example, 5S with standard material locations and SMED in tool exchange is used out of the lean toolbox. A strong CI culture is present with management commitment to explore and exploit efficiency gains through automation and employee involvement. On top of that, a dedicated innovation team is in place which optimizes current processes through continuous improvement and explores new technologies on a full-time basis. However, there is room for improvement in scheduling orders, which is currently done manually. Additionally, partnerships in the supply chain could be improved in for example staining. Currently, some stock is kept to fulfill some customer orders within the contractual 48-hour timeframe as the staining company cannot deliver JIT. Overall, this company is scaled in phase 4 of the lean maturity and pursuing lean management perfection.

Regarding smart manufacturing, they are currently entering stage 2 by piloting in for example cobots. Furthermore, there are two track and trace systems that scan and print QR-codes, two welding robots, an ERP system that is constantly being updated and extended by system integration, a camera vision system for quality checks, and a bending machine that automatically changes to the correct tooling for a specific drawing.

(22)

being extended with extra modules. For example, personalized shop floor control for each work bench screen that only provides relevant information (e.g. 2D drawings or 3D models) for a specific job or employee and furthermore indicates schedule priorities. Hence, there is no need to contact office employees or customers for information. All information is easily accessible for the shop floor and no time is lost due to information inertia. Fourth, a camera vision system is being used for quality reports that are being required by customers and thus adds value in a short time. A hand-held device is used to touch the product and quickly check whether or not it is in line with product specifications. Additionally, these quality checks can quickly provide feedback to the shop floor for continuous improvement. Fifth, a bending machine is being used with automatic tool exchange according to the SMED principle. It creates a better flow, improves quality for customers by being error-proof in using the correct tool and installation for the product (i.e. standardization), and it allows for other value adding activities for a shop floor employee during tool exchange. Sixth, a pilot was running with a cobot that automated material placement in a bending machine. This allows for more precise and standardized bending and possible cost improvements. Seventh, a track and trace system (QUMA-machine) was installed to increase traceability by scanning plates for identification and accordingly assigning operations in ERP and printing QR-codes. However, this system is currently not being used anymore due to difficulties in testing and management commitment. In general, the observations exemplify that SM and LM can complement each other and jointly increase performance.

4.1.3 Company C

(23)

flow in the operations and reduce huge waiting times due to internal logistics and office operations. Additionally, there is information inertia between the shop floor, the office and the customer. In short, the lean maturity is in phase 2 due to the identification of value streams.

“Our employees are only able to perform operations for a specific function. Take for example nesting, if there is nothing to nest they are simply idle. However, the person who is preparing the work is heavily loaded. Once the work reaches the nester, an enormous amount of work has to be processed which he cannot process all at once. Hence, huge waiting times occur.” (Company C)

In the current state company C is thus willing and starting to implement lean principles. Hence, smart manufacturing is not of the highest priority, immature and very slowly starting to emerge. Base technologies are not present and thus only little options are available to create a smart factory. Contrastingly, the need for these technologies is implied by improving predictability of lead times and more flexibility of the organization. Nonetheless, a few smart technologies are identifiable: an ERP system is in place and currently being updated, there is an integration of systems to share drawings, nesting operations are semi-automated, and a pilot in 3D scanning the shaped metal plates. As a consequence, smart maturity is low, slowly emerging and thus stage 1 (vertical integration).

(24)

interrupt flow. All in all, the smart technologies are again used to increase flows of information and to quickly perform quality checks within the station. However, exemplified by the nesting automation, one has to be careful what and how smart technologies are being exploited to create flow as it may quickly become anti-lean.

4.1.4 Company D

The philosophy of the fourth case company is revolving around modularity in design. As the shop floor is mainly performing assembly tasks, they benefit from standardization in product design (i.e. modularity). This translated into lower stock levels and being able to procure at a higher level of the bill of materials from low-wage countries, which reduced both costs and internal cycle times. Consequently, due to mainly modularity, a flow of common products was established on the shop floor and structure was brought into the processes (e.g. 5S and work cells). Together with the desire to reduce cycle times according to the QRM philosophy, a cycle time reduction of 8 weeks was realized. In general, the lean maturity is scaled in phase 4.

A major cycle time reduction is not only established by standardizing processes and design. Automation (e.g. Jidoka) or smart manufacturing technologies have brought major improvements on office operations. For example, shop floor control, planning automation, and mobile logistics are being exploited. Hence, smart manufacturing maturity is in the beginning of stage 2 (automation). A key statement of the managers is that they use smart to become lean. This clearly indicates that company D exploits smart technologies that fit into the lean philosophy.

“Smart and lean are closely related and interlocked, because we use smart to become lean and reduce waste. That is the fun part of smart and lean, it strengthens each other. A new smart technology can pull a process to a higher level.” (Company D)

(25)

competencies and production priorities. Hence, repetition in tasks is reduced. However, some prerequisites had to be met. For example, in order to fully exploit shop floor control in the ERP system, planning should be performed in the ERP system in a standardized way, bill of materials needed to be updated, and work instructions had to be digitally available and accurate.

Second, the (service) planning is made automatically in ERP. Hence, no time is lost on complex manual planning activities in Microsoft Excel and transferring data from system to system. The ERP system provides a best schedule after which a planner can choose to make some deviations. Again, in order to implement this smart technology some prerequisites had to be met. The planning should be made in the ERP system, and billing and the stock handling had to be automated. For example, billing is now done instantly with a push of a button. The pdf is generated and automatically sent to the customer. No need to waste time on manually generating the pdf, writing an e-mail and attaching the pdf.

Third, mobile logistics are being used in the warehouse. A material runner uses a hand-held device that provides information for internal or external orders. This means that stock has standard locations, less time is wasted on searching for materials and monitoring performance is automated. Hence, it becomes easier to track where time is lost and thus what operations could be improved by means of continuous improvement. Moreover, the operations manager stated that it became a game for the material runner to perform a specific operation within the hour instead of half a day.

The beforementioned descriptions of mechanisms clearly show that SM technologies are being used to become more lean. Moreover, SM can even become a driver for lean. Thus, smart manufacturing can pull lean operations to a higher level.

4.2 Success factors

This section is identifying factors that can influence the level of interactions between LM and SM. First, we elaborate on the role of management. Second, the influence of available or allocated resources is described. Finally, the importance of knowledge will be analyzed.

4.2.1 Management

(26)

First and foremost, there needs to be a vision of the interplay between lean and smart manufacturing. A common vision of the management team on how to maximize customer value by exploiting lean and smart manufacturing is thus required. This vision should subsequently be translated into how processes are going to be structured and monitored. On the basis of vision and process structure one can choose the automation that is required to achieve the desired performance. However, half of the case companies (A&C) lack a vision with regard to smart manufacturing. Company B and D elaborated on a clear vision for their current and future state.

“In my opinion it is very important to have vision for smart industry or the importance of IT in the future of a company. However, we do not have a vision yet. The opinions in the management team are divided. IT is something incredibly nontransparent and hardly anyone fully understands.” (Company A)

Second, after reaching a vision management needs to commit to LM and SM to be integrated into the organization. Without management knowledge of smart manufacturing and management commitment to exploit lean and smart manufacturing and to support it throughout the organization, the implementation will evidently fail. This is further supported by the example observed at company B regarding the tracking and tracing system of the QUMA, which automatically scans and detects products, prints a QR-code and assigns routes to the materials. The machine was bought without full consensus of the management and thus had little management support. Furthermore, they needed to create partnerships with the machine producer and a programmer, which created difficulties in collaboration. Moreover, the pilot had problems with lost materials and hence the management support level fell down. Consequently, they stopped using the machine as it costed too much resources and the potential performance gain is lost. All in all, Company A and C are committed to lean but lack commitment to combine lean and smart manufacturing due to the lack of a vision. In contrast, company B and D are committed to combine lean with smart.

(27)

workforce. Company A, B and D expressed a decent level employee involvement although the latter wants to improve by implementing A3 problem solving out of the lean toolbox.

“In order to make SM work at the shop floor level, management has to keep repeating its importance … That is something where we [management] have to be in alignment and committed to. If I am the only one supporting SM, it becomes a mission impossible. If employees dug in the heels nothing will happen.” (Company C)

To compare the effect of management, consisting of the subfactors vision, commitment and employee involvement we plot the level of interactions between lean and smart. What can be distilled from figure 4 is that a high level of interactions (i.e. combined lean and smart maturity) is enabled by a high level of management support.

Figure 4, Interaction level of LM & SM vs management matrix

4.2.2 Resources

The importance of allocating resources in financials, employees and time was indicated by the example of company B regarding the failure to implement the QUMA. Nevertheless, company B has a dedicated innovation team and has successfully implemented for example welding robots and QRM. Two employees are working full-time on improvements on process improvements and constantly investigate purposes for new smart technologies. However, this was not common practice in the past.

(28)

In comparison, company A, C, and D do not have a dedicated team. Company A and D assign temporary project teams with stakeholders from relevant departments for such improvements. They execute these projects as a side job and can thus not fully dedicate their time for these projects. More importantly, it is thus not their highest priority. However, company D managed to successfully implement shop floor control, digital warehouse logistics and automated planning by using short-term teams. They are willing to allocate resources to the team and partnerships (e.g. consultants), while company A is unable to allocate such an amount of resources to these projects. Company C showed no project teams for smart manufacturing. They stated that they are mainly allocating resources to implement QRM.

In general, we observed a higher lean and smart manufacturing maturity at companies that dedicated resources for integration projects. The four companies state it is a trade-off between resources and profits or knowledge and thus it is simply a business case. However, they all mention that there is always a lack of resources in both financials and employees (e.g. lean and smart specialist or job specific specialist).

“We only have 35 employees. Quite quickly we consult people to execute integration projects for us. We cannot do it all ourselves” (Company D)

(29)

4.2.3 Knowledge

Another factor that is impacting the level of interactions mechanisms between lean and smart is knowledge. It is by far the most observed factor (27 codes) from the interviews (refer to

Appendix C for a summary of the code table). We observed that there is a lack of knowledge

on especially smart manufacturing throughout the four different case companies. Low-threshold or easy to use and understand techniques like smart planning or shop floor control were observed. However, more cutting-edge techniques like Virtual Reality or 3D printing lack practical presence. The immaturity is explained by the fact that all companies struggle to find purpose or knowledge on how to exploit the wide variety of smart techniques combined with lean principles.

“The theoretical basis for applications within our production site is often missing” (Company B)

Additionally, we observe that although the main ideas of lean and smart might be clear, it is quite hard to translate, adjust and exploit them in a case-specific environment. Let alone, the knowledge to exploit both together in their greatest potential. Therefore, partnerships were observed in all four cases to in-source the required knowledge.

“We have enough technical knowledge for the product itself. However, regarding LM and SM projects we lack knowledge. That’s why I am very happy with the support of the universities … You have to be able to make it fit in your own company. That’s very tricky.” (Company C)

(30)

Figure 6, Level of interaction LM & SM vs Knowledge matrix 4.3 Future state

Not only the current state was discussed with regard to lean and smart manufacturing. Several use-cases or ideas have been discussed where lean and smart could interact with each other. Therefore, this section focusses on the desired or future state.

4.3.1 Company A

As previously elaborated, company A has problems to deal with the wide variety of products in the production planning, which creates long cycle times or waiting times. The complexity is simply too much to oversee for a human or to be flexible to reschedule for contingencies. A smart planning algorithm could easily reduce this complexity by calculating and proposing the best possible schedule visually. Hence, flow in materials can be increased by finding the most efficient sequence in which set-up times are being minimized. Furthermore, they expect that smart manufacturing (e.g. big data analytics) could recognize the most efficient mix (e.g. schedule) in make-to-stock and make-to-order to establish a more efficient machine utilization. The repetitive nature of make-to-stock should support make-to-order production. This could possibly level demand internally, create flexibility and a constant flow.

(31)

Additionally, potential was recognized in automated guided vehicles (AGV’s), augmented reality (AR) and information dashboards. The AGV would find purpose in reducing travel time by humans and the augmented reality glasses was proposed to be used in a warehouse picking system. It could indicate the best possible route and visually show the locations and quantities to be picked thus increasing flow. Additionally, information dashboards are useful to provide employee with relevant standardized work information or analysis of the production to steer on key performance indicators of the value stream. This would ease the extraction of gathered ERP data and support analysis of this data that is being visualized on the dashboard to steer on key performance indicators. Hence, at each or the lowest hierarchical level opportunities can be identified to continuously improve their own work.

4.3.2 Company B

Although company B was observed as the most mature in smart manufacturing, they see plenty of room for improvement with regard to smart manufacturing. First, they also recognize the potential of smart planning. It would reduce the complexity of capacity planning on the shop floor by quickly providing options for an ideal or most efficient planning. Hence, the flow of products could be improved by minimizing set-up times and thus reducing waiting times.

Second, they also recognize the potential of ‘chain thinking’. However, they want to take it one step further: they see opportunities for immediate confirmation by a smart technique that recognizes the materials, required routes, cycle times, delivery date and price on the basis of the uploaded drawings. The real-time confirmation would be in line with the lean principle to have a direct customer link with a real-time ‘yes’ or ‘no’ signal for request and responses (Spear and Bowen, 1999).

Third, they are quite sure that AGV’s could potentially reduce non-value adding motion and transportation times of shop floor employees. This would not only increase the flow of materials, but also in information. An example was provided in an IoT system: “the cobot gives

a signal to the ERP that the product is ready, subsequently the AGV drives to the cell, retrieves the materials and brings it to the next location. Meanwhile, the ERP commands the automatic tool exchange machine to change tools for the next order.” (Company B)

(32)

The tablet was also mentioned as a more flexible form of working on screen or shop floor control. Currently, the screens are fixed and time is lost in comparing drawings with the physical product on the work bench. Additionally, it could act as a dashboard that only presents relevant information for the specific user. A shop floor employee should see drawings and quality checks, while a supervisor would be able to monitor key performance indicators in real-time exploiting big data analytics.

Finally, they want to make another effort to make the QUMA track and trace system work. The potential of reducing wait times a by detecting raw materials, assigning QR-codes and the automatic assignment of routings is still promising to reduce wait times of raw materials and thus improve cycle times, flow and traceability.

4.3.3 Company C

Previously, in section 4.1.3 we observed that this company is struggling to forecast and level their operations due to the high fluctuations in demand. They would like to account for this irregular demand by being flexible in the workforce. Smart or semi-automated tools could reduce the complexity of operations. For example, the forecasting of required time through a smart system could prevent the mismatch in calculations and thus establish clarity in go or no-go signals to customers. In collaboration with a university student they are creating a forecast model that is able to provide a more precise forecast on required routes, employee competencies and time.

Another topic is tracking and tracing of the products to improve process transparency. Currently, they identified that long waiting times occur due to the gathering and movement of material in batches on big trailers. These trailers can be stationary for at least a couple of days but they are unable to monitor the problem on a raw material level. Hence, by integrating a tracking and tracing system they might be able to identify the problem or to connect it with a pull signal to increase flow.

(33)

4.3.4 Company D

Similar to company B, which has the same maturity in lean and smart, chain thinking is a hot topic for company D. Real-time information sharing could trigger or send signals to customers or supplier to improve direct customer linkage and improve supply chain integration. What is more, this company has the desire to add an online configuration tool. Within the boundaries of the modularity, product can be configured and ordered in real-time to speed up order confirmation and improve transparency. Additionally, tracking and tracing in the supply chain could be improved by using RFID chips on products. They are more reliable compared to for example QR-codes in case of dirty or damaged that make the QR-code unreadable. The radio signal can be automatically detected and thus improve traceability.

(34)

5. Discussion

In the following section, we first discuss the main findings and how they relate to current theory. Second, practical implications of the main findings are being provided. Third, limitations of this will be identified and discussed.

5.1 Main findings

A commonality across all cases is that relatively low-level or passive smart manufacturing technologies are currently being used to reduce information inertia. Most companies recognize the potential of smart manufacturing but struggle to start or integrate SM in their processes. The complexity of operations in a high variety and low-volume environments poses difficulties in automating or standardizing the wide range of processes. Stage 1 of Frank et al. (2019) smart implementation framework is the most common, whereas one case is giving priority to first implement lean principles prior to starting with SM, while another is on the frontier with stage 2 piloting in cobots.

Currently, vertical integration through smart is mostly being used in the companies (e.g. ERP, shop floor control, tracking and tracing). The informational systems of different hierarchical levels are interconnected through the ERP system. This interconnectivity speeds up information sharing within the organizational boundaries (Moeuf et al., 2017). Hence, information inertia is reduced and transparency increased. Horizontal integration is a subsequent stage of smart implementation, which is in line with the proposed LM and SM integration framework of Sony (Sony, 2018) and Frank et al. (2019). However, it is mainly being mentioned as a desired future state to standardize and share real-time information throughout the supply chain (Duarte, 2017).

Above all, companies are reducing waste and increase performance by automating manual and repetitive processes in mainly office tasks (e.g. production planning), which is line with the lean principles like work standardization and the creation of flow. Hence, smart manufacturing can support lean management (Dombrowski, Richter and Krenkel, 2017; Sanders, K. Subramanian, et al., 2017).

(35)

processes need to be structured and thus become lean. Thus, the implementation or exploitation of smart manufacturing technologies act as a magnifying glass to identify badly structured processes (Bortolotti and Romano, 2012). However, this is beneficial in identifying non-value adding processes. On the other hand, company C showed that implementing lean was of a higher priority than smart. Therefore, it is still beneficial to first have a foundation of lean prior to start with exploiting smart manufacturing technologies.

Additionally, smart manufacturing can complement lean in continuous improvement. For example, performance monitoring is easier with the use of sensors and digital production data measurements. Moreover, the big data can be analyzed and identify opportunities for process improvement. Hence, we acknowledge that “the fundamental idea of continuous improvement is no longer purely connected to the human being” (Tsipoulanidis, 2017). In fact, smart technologies like big data analytics can be used to for example fill the boxes in the lean tool A3 problem solving.

Regarding the organizational context and its success factors management, resources and knowledge, they are as expected in line with lean management critical success factors (Bortolotti, Boscari and Danese, 2015; Valaei and Rezaei, 2016; Almanei, Salonitis and Xu, 2017). Moreover, the patterns in figure four, five and six are similar which further support that these factors correlate with each other. Management choices including vision, commitment, knowledge and employee involvement do positively correlate with the amount of interaction mechanisms between lean and smart. These top-down management choices subsequently translate into the amount of resources allocated to support these projects. The more resources are being allocated, the higher the interaction level between lean and smart. However, due to the fact that SMEs were observed, we have to note that these resources are limited and therefore might be immature (Moeuf et al., 2017). Finally, knowledge on smart manufacturing is in general low to medium. Hence, knowledge of how to combine lean and smart manufacturing is low. This lack of knowledge explains the rather low maturity of interactions between LM and SM observed in the case companies.

(36)

“The lack of digitalization is restraining humans to understand the complexity in high variety operations … but maybe especially in that complexity one could exploit big data to see the bigger picture.” (Company A)

5.2 Practical implications

This case study presented some relevant implication for practical purposes. First, the implementational frameworks of lean can be adapted for smart implementation or ultimately joint implementation. Three critical success factors will positively influence the integration of LM and SM: (1) Management: which first needs to have sufficient knowledge of LM and SM. Subsequently, this knowledge has to be translated into to a vision on how to maximize customer value by exploiting lean and smart jointly together. Furthermore, commitment is required to keep supporting and creating a culture that is willing to participate in the change; (2) Resources: this closely interrelates with the commitment, without allocating a substantial amount of resources to execute LM and SM integration project, the improvement project have a high chance of failure. Not only financials are required, but also time (from employees or partners) for successful implementation; (3) Knowledge: which is created or acquired by running pilots or creating partnerships.

(37)

6. Conclusion

In this final section, conclusions are drawn on the main findings to answer the research question. Subsequently, the limitations are identified and discussed together with suggestions for further research.

6.1 Answering the research question

The main research question of this research was: how are underlying mechanisms of smart and

lean influencing operational performance? By observing four different case companies we

created a better understanding on how to jointly exploit LM and SM. We conclude that SM and LM can support and complement each other in a high variety and low volume environment. First, the interaction mechanisms increase the flow of information through vertical integration and thereby enable an increase in flow of materials through standardization, a reduction in complexity, and the automation of repetitive tasks. Furthermore, smart can improve process transparency, traceability, and identify or reduce non-value adding activities in order to maximize customer value. Thereby, SM is able to enhance the speed of lean principles like continuous improvement, performance monitoring, and total quality management. Second, horizontal integrations establish real-time information sharing in the supply chain. Hence, signals for pull are send in real-time to align operations throughout the supply chain. In addition, similar to LM, management, resources, and knowledge are critical success factors to establish value-adding interactions.

6.2 Limitations and future research

Reflecting on the research method, several limitations can be observed. First, as smart manufacturing is quite immature in practice, it was hard to identify state-of-the-art or highly complex smart technologies in close interaction with LM. In addition, as both academics and managers are struggling with definitions or integrations of SM, it is hard to theorize exactly how interactions mechanism between LM and SM work. Moreover, the fact that we observed SMEs may even explain the low maturity as they typically have less resources available (Pearce, Pons and Neitzert, 2018). Hence, we strongly suggest to research big enterprises to have a better chance of identifying state-of-the-art smart technologies.

(38)

high variety of tasks, while still staying flexible. Furthermore, this probably changes how smart technologies can be applied and thus how they interact. Not to mention the fact that different markets were studied within the same process type. Consequently, the influence of the process type and market on the applicability and performance of smart manufacturing could be researched.

Third, the qualitative research was conducted to explore interactions mechanism and this requires a descriptive approach. Therefore, we had to be careful to make major conclusions and rather provide propositions. The gathered data through semi-structured interview is highly reliant on opinions or perspectives of the interviewees. A quantitative study should be conducted to measure and prove the increase in performance statistically.

(39)

References

Ahuja, I. P. S. and Khamba, J. S. (2008) ‘Total productive maintenance: Literature review and directions’, International Journal of Quality and Reliability Management, 25(7), pp. 709–756. doi: 10.1108/02656710810890890.

Almanei, M., Salonitis, K. and Xu, Y. (2017) ‘Lean Implementation Frameworks: The Challenges for SMEs’, Procedia CIRP. The Author(s), 63, pp. 750–755. doi: 10.1016/j.procir.2017.03.170.

Alves, A. C., Dinis-Carvalho, J. and Sousa, R. M. (2012) ‘Lean production as promoter of thinkers to achieve companies’ agility’, Learning Organization, 19(3), pp. 219–237. doi: 10.1108/09696471211219930.

Anvari, A. et al. (2011) ‘A proposed dynamic model for a lean roadmap’, African Journal of

Business Management, 5(16), pp. 6727–6737. doi: 10.5897/AJBM10.1278.

Bhamu, J. and Sangwan, K. S. (2014) ‘Lean manufacturing: Literature review and research issues’, International Journal of Operations and Production Management, 34(7), pp. 876–940. doi: 10.1108/IJOPM-08-2012-0315.

Bortolotti, T., Boscari, S. and Danese, P. (2015) ‘Successful lean implementation: Organizational culture and soft lean practices’, International Journal of Production Economics. Elsevier, 160, pp. 182–201. doi: 10.1016/j.ijpe.2014.10.013.

Bortolotti, T. and Romano, P. (2012) ‘“Lean first, then automate”: A framework for process improvement in pure service companies. A case study’, Production Planning and Control, 23(7), pp. 513–522. doi: 10.1080/09537287.2011.640040.

Buer, S. V., Strandhagen, J. O. and Chan, F. T. S. (2018) ‘The link between industry 4.0 and lean manufacturing: Mapping current research and establishing a research agenda’,

International Journal of Production Research. Taylor & Francis, 56(8), pp. 2924–2940. doi:

10.1080/00207543.2018.1442945.

Chay, T. et al. (2015) ‘Towards lean transformation: the analysis of lean implementation frameworks’, Journal of Manufacturing Technology Management, 26(7), pp. 1031–1052. doi: http://dx.doi.org/10.1108/09564230910978511.

Dombrowski, U., Richter, T. and Krenkel, P. (2017) ‘Interdependencies of Industrie 4.0 & Lean Production Systems: A Use Cases Analysis’, Procedia Manufacturing. The Author(s), 11(June), pp. 1061–1068. doi: 10.1016/j.promfg.2017.07.217.

Duarte, S. (2017) ‘An investigation of lean and green supply chain in the Industry 4 . 0’,

(40)

265.

Dul, J. and Hak, T. (2007) Case Study methodology in business research, Case Study

Methodology in Business Research. doi: 10.4324/9780080552194.

Flynn, B. B., Schroeder, R. G. and Sakakibara, S. (1994) ‘A framework for quality management research and an associated measurement instrument’, Journal of Operations Management. doi: 10.1016/S0272-6963(97)90004-8.

Frank, A. G., Dalenogare, L. S. and Ayala, F. N. (2019) ‘Industry 4.0 technologies : Implementation patterns in manufacturing companies’, Intern. Journal of Production

Economics. Elsevier B.V., 210(January), pp. 15–26. doi: 10.1016/j.ijpe.2019.01.004.

Hasle, P. et al. (2012) ‘Lean and the working environment: A review of the literature’,

International Journal of Operations and Production Management, 32(7), pp. 829–849. doi:

10.1108/01443571211250103.

Hermann, M., Pentek, T. and Otto, B. (2016) ‘Design Principles for Industrie 4.0 Scenarios’,

49th Hawaii International Conference on System Sciences (HICSS). IEEE, pp. 3928–3937. doi:

10.1109/HICSS.2016.488.

Hines, P., Holweg, M. and Rich, N. (2004) ‘Learning to evolve’, International Journal of

Operations & Production Management, 24(10), pp. 994–1011. doi:

10.4324/9781315857817-11.

Hof, R. M. (2017) ‘The Synergy between Lean Principles and Smart Manufacturing Technologies The Synergy between Lean Principles and Smart Manufacturing Technologies’. Holweg, M. (2007) ‘The genealogy of lean production’, Journal of Operations Management, 25(2), pp. 420–437. doi: 10.1016/j.jom.2006.04.001.

Kagermann, H., Wahlster, W. and Helbig, J. (2013) ‘Recommendations for implementing the strategic initiative INDUSTRIE 4.0: Final report of the Industrie 4.0 Working Group’, Final

report of the Industrie 4.0 WG, (April), p. 82. doi: 10.13140/RG.2.2.14480.20485.

Kang, H. S. et al. (2016) ‘Smart manufacturing: Past research, present findings, and future directions’, International Journal of Precision Engineering and Manufacturing - Green

Technology, 3(1), pp. 111–128. doi: 10.1007/s40684-016-0015-5.

Karlsson, C. (2016) Research Methods for Operations Management, Research Methods for

Operations Management. doi: 10.4324/9781315671420.

Kolberg, D. and Zühlke, D. (2015) ‘Lean Automation enabled by Industry 4.0 Technologies’,

IFAC-PapersOnLine, 48(3), pp. 1870–1875. doi: 10.1016/j.ifacol.2015.06.359.

Exploring underlying mechanisms influencing Smart & Lean manufacturing performance: a multiple case-study