Identifying relationships between the company focus, facility layout and product architecture

(1)

Identifying relationships between the company focus, facility layout and product architecture

8/14/2011 Klaas Werkhoven S1830740

1

Identifying relationships between the company focus, facility

layout and product architecture

University of Groningen Faculty of Economics and Business

Technology Management Master Thesis

Author: Klaas Werkhoven

(2)

2

Identifying relationships between the company focus, facility

layout and product architecture

Klaas Werkhoven

University of Groningen

Faculty of Economics and Business

Technology Management

14 August 2011

Meester Klaas Douwesstraat 4 8804 NW Tzum

(3)

3

Reading advice

Because not every person might be interested in reading this report completely, due to time restraints or specific interests, this reading advice is added to advice the reader on which parts of the report to read.

Reading advice based on available time.

10 min – If there is only limited time available the reader is advised to read the abstract (page 4) and the conclusion (page 53).

20 / 30 min – If there is approximately 20 to 30 minutes available for reading this report the readers is advised to read the abstract (page 4), analysis and results (page 42) and the conclusion (page 53).

40 / 60 min – If there is approximately 40 to 60 minutes available for reading this report the reader is advised to read the abstract (page 4), literature review (page 14), conceptual model and hypotheses (page 25), analysis and results (page 42) and the conclusion (page53)

60 min + The reader is advised to read the entire report.

Reading advice based on the interests of the reader.

Literature If the reader is interested in the literature available on the research topics of this thesis, the reader is advised to read the literature review (page 14) and conceptual model and hypotheses (page 25).

results If the reader is interested in the results of this research then the reader is advised to read the abstract (page 4), analysis and results (page 42) and the conclusion (page 53).

(4)

4

Abstract

This research is aimed at identifying the relationships between the company focus and the corresponding production performance measures, the facility layout, the product architecture and the firm response to the market. The company focus as well as the product architecture and facility layout are well researched topics, however up until now only these specific topics are researched, and not their relationships. In order to overcome sub-optimizations it is necessary to investigate the relationships between these various specific research areas.

Three hypotheses and corresponding models are created based upon literature about these topics. Each of the hypotheses is focused on one specific customer focus (customer intimacy, operational excellence or production performance). Each customer focus leads to a set of performance measures, in order to make sure that a specific strategy is followed. The product architecture and facility layout are tuned to facilitate the company specific strategy.

(5)

5

Preface

As a final demonstration of competence each student from the master program Technology Management at the faculty of Economics and Business at the University of Groningen (RUG) has to write his or her master thesis. The research has to be performed under the supervision of a researcher of the RUG, and has to focus on a predefined specific area of interest.

This report describes the research I did and focus on the relationships between product architecture, facility layout and the company focus. To focus my thesis on this subject is because I have a high interest in the daily operations of production companies, which is a often discussed subject in the courses of the master and bachelor Technology management. Furthermore for my bachelor thesis I investigated which factors influenced the success of Lean manufacturing. Because of my interest in the daily operations of production companies I chose to investigate the relationships between product architecture, facility layout and the company focus. By choosing this topic the opportunity exists to visit multiple companies, instead of only one when performing a company related assignment. This, in my opinion, makes the assignment more difficult, but can be a good contribution to my learning process.

Lastly, I want to thank and show my appreciation to my first supervisor dr. L. Zhang for the good advices during this research project, for the always very quick responses, clear answers to all my questions, and for bringing me in contact with one company. Next to dr. L Zhang I would like to thank prof. dr. ir. J. Slomp for bringing me in contact with several companies. Furthermore I would like to thank dr. N.D. van Foreest for being my second supervisor.

(6)

8/14/2011 Klaas Werkhoven S1830740 6

Index

Reading advice ... 3 Abstract ... 4 Preface ... 5 1. Introduction ... 8

1.1. Company focus, product architecture and facility layout ... 8

1.2. Problem statement ... 9 1.3. Objective ... 10 1.4. Research question ... 11 1.5. Scope ... 11 1.6. Report structure ... 13 2. Literature review ... 14

2.1. Company focus and related production performance ... 14

2.2. Product architecture ... 17

2.3. Different types of facility layout ... 20

3. Conceptual model and hypotheses ... 25

3.1. Relationships between company focus and production performance measures ... 25

3.2. Relationships between production performance measures and product architecture 28 3.3. Relationships between production performance measures and facility layout ... 31

3.4. Relationships between product architecture and facility layout ... 33

3.5. Hypotheses ... 36

4. Research design ... 39

4.1. Research method ... 39

4.2. Case selection ... 39

4.3. Data collection and instrument development ... 41

5. Analysis and results. ... 42

5.1. Analysis of elements ... 42

5.2. Bilateral relationships ... 47

5.3. Multilateral relations ... 50

6. Conclusion ... 53

6.1. Limitations ... 55

6.2. Suggestions for further research ... 56

References ... 57

Books ... 57

(7)

7

Websites ... 62

Appendix I : Considerations research design ... 63

Appendix II: Questionnaire ... 67

Appendix III: Validity of the data ... 91

Appendix IV: Considerations instrument development ... 92

List of tables, figures and abbreviations ... 95

(8)

8

1. Introduction

Companies are always trying to perform as good as possible, however the standard for “good performance” is always changing, and thus companies have to change. Currently organizations are driven by more demanding customers and more global competition (Woodruff 1997). To keep a competitive advantage companies are often internally oriented and are trying to make performance improvements (Woodruff 1997).

1.1. Company focus, product architecture and facility layout

Often performance measures are used to determine if a specific goal or target is met. A statement often made regarding the use of performance measures is: “What gets measured gets done”. However according to Behn (2003) “what people measure often is not what they want done. And people – responding to the explicit or implicit incentives of the measurement will do what people are measuring, not what these people actually want done”. Because of this it is necessary for companies to know what it is they want to achieve, and how they want to achieve this. Often companies focus on the customer to determine their strategy. Companies which focus on the customer often make use of one of the three value disciplines described by Treacy & Wiersema (1993). The three value disciplines described by Treacy & Wiersema (1993) are customer intimacy, operational excellence and product leadership. When trying to provide the customer with the product he desires companies have to make multiple decisions. One of the important decisions a company has to make are decisions regarding the product architecture. If for example a customer wants a product produced to his specific specifications the product architecture should be arranged accordingly. Furthermore some researchers claim that it is of importance to match the product architecture and the facility layout. Erixon (1996) for example states: “when a product is modularized, the factory should be reorganized accordingly”.

(9)

9 production costs, or to improve “global performance characteristics”. However in other cases a modular architecture can also be used to reduce costs, or to increase the number of different products a company can offer. Each of these decisions can influence the way value is delivered to the customer.

The facility layout can also have a large impact on the production performance of a company. However as many authors mention, capturing the relationships between operational performance and a layout configuration is notoriously difficult (Benjaafar et al., 2002 & Yang, 2000). Historically the evaluation criteria which was mostly used to evaluate the conventional layouts is the amount of material handling (Nicholas, 2008). If only one product is produced in a repetitive manner, the material handling can be reduced relatively easy by placing consecutive operations in close proximity, which results in a product layout. However if multiple products are produced which each have different routings, this could become difficult. Next to this problems can occur if the volume changes or if the product mix changes regularly. Some researchers claim that it is of importance to match the product architecture and the facility layout. “When a product is modularized, the factory should be reorganized accordingly” (Erixon 1996).

1.2. Problem statement

The previous mentioned examples of specific research fields (e.g. product architecture or facility layout) are all focused on one specific topic. Based on the previous introduction it becomes clear that much research is already done on specific topics in order to ultimately improve the production performance of companies, which in turn has to lead to increased company performance. However, the majority of the research is focused on a specific topic, which can defended from a reductionist point of view. However, this doesn’t necessarily have to lead to an optimal solution. The reason for this is that complex systems can exhibit emergent properties (Jackson, 2000).

(10)

10 architecture and no variations of the product are made, in this case a delayed differentiation layout would make no sense and could be far from optimal.

In this research the relationships between the selection of production performance measures and their relationships to the facility layout and product architecture is examined. Product architecture as well as facility layout can have large impacts on the production and the overall performance of companies (Ramdas 2003), thus it is necessary to better understand the relationships between these different strands of research. This has to prevent companies from making changes in the product architecture or facility layout that seem beneficial but ultimately prove to be less successful than intended.

1.3. Objective

From the introduction and the problem statement it becomes clear that, although it is necessary and useful to research specific topics, it is also necessary to investigate the relationships between various research topics. As mentioned in the problem statement this research project focuses on how the company focus on its customers and the corresponding selection of production performance measures which affect the facility layout and product architecture.

Main objective

To identify relationships between the company focus, product architecture and facility layout.

In order to accomplish the main objective, it is divided into multiple sub-objectives, each of these sub-objectives is elaborated on in the following parts of this report.

Sub-objective 1

Develop a conceptual model of the relationships between the company focus, product architecture and facility layout based on current available literature.

Sub-objective 2

(11)

11

Sub-objective 3

Test the hypotheses with multiple cases in order to show the potential for further research.

1.4. Research question

The combination of the information from the problem statement, objective and the scope of the project leads to the following research question.

Research question

What are the relationships between the company focus, product architecture and the facility layout?

To answer the research question it is divided into four sub-questions. The answers on these sub-questions lead to a detailed answer of the main research question.

Sub-question 1

What are the relationships between the company focus and the selection of production performance measures?

Sub-question 2

What are the relationships between the selection of production performance measures and the product architecture?

Sub-question 3

What are the relationships between the selection of production performance measures and the facility layout?

Sub-question 4

What are the relationships between the product architecture and the facility layout?

1.5. Scope

(12)

12 proposed hypotheses. The theoretical scope sets the boundaries for the topics that are researched.

Project scope

As explained the main objective for this research is to identify relationships between the company focus, product architecture and facility layout. To achieve this objective within the available time to perform this research, the scope of this research is limited to some degree. The development of a conceptual model and hypotheses based on current available literature, gathering of the data for the verification of the hypotheses are within the scope of this project. However because there is limited time available to investigate the relationships between the company focus, product layout and facility layout at different companies, only a small number of companies is visited. The actual large scale data collection and processing of the collected data to verify the hypotheses is beyond the scope of this project.

Theoretical scope

Both product architecture as well as the facility layout are well researched topics. To explore how every specific researched strand within these topics has its own effects, would be too much within the available time. For this reason it is chosen to focus on different types of product architectures and different types of facility layouts.

The different types of product architectures which are examined are different forms of modular architectures and an integral architecture. Furthermore layouts ranging between these two different types of architectures are included. The manner in which an architecture is established, and how this affects the facility layout or product performance is also be within the scope of this project. The manner in which an architecture is established is included, because if for example, the four step method described by Ulrich & Eppinger (2008) is used, the clustering of elements is partially based on capabilities of vendors. These specifics capabilities could originate from a specific facility layout at the vendor, however this could also be true for internal departments which provide elements for the final product.

(13)

13 problem. Such algorithms determine which type of layout will be used or specific placements of machines or departments within a layout type. However the number of specific ways to arrange machines or departments within a layout type is unlimited, for this reason this research focuses on the types of layout used by companies. The layout types that are examined range from more traditional factory layouts (e.g. functional or cellular layout) to newer types of factory layouts (e.g. Spine layout, star layout or distributed layouts) (Benjaafar et al., 2002).

1.6. Report structure

(14)

14

2. Literature review

This chapter presents the theoretical background for the conceptual model of the relationships between the company focus and the corresponding selection of production performance measures, product architecture and the facility layout. For each topic the most relevant literature for this research is discussed.

2.1. Company focus and related production performance

A fast amount of literature is available on the topic performance measurement and specifically the performance measurement of production environments. However despite the fast amount of literature available on the performance of production systems it is hard to construct a general theory on how the production should be enhanced by companies. The reason for this is that performance measurement systems are often multidimensional, and what is perceived as performance can differ per company, because the performance has to be in line with a certain strategy in order to be of significance. The multidimensionality of production performance and the strategy related to production performance is discussed in the next two paragraphs.

Multidimensionality of production performance

The literature on multidimensional performance measures provides an array of different models raging from 3 dimensions (Pettersen et al., 2010) to 22 dimensions (Avella & Vázquez-Bustelo, 2010) and even up to 184 dimensions for world class manufacturing (Digalwar & MetrI, 2005). Pettersen et al., (2010) for example analyze three “basic” performance measures, work in progress (WIP), lead time and throughput. Avella & Vázquez-Bustelo (2010) use 22 measures of performance which they classify as “manufacturing objectives / priorities”, each of which they place in one of the five performance categories they identified (cost, flexibility, quality, delivery and environmental protection). Especially the models which use a multitude of measures to determine the performance make use of categories and even sub-categories to build a performance model. De Toni & Tonchia (2001) for example use two main categories, cost and non-cost which they use to place the sub-categories production and productivity costs and time, flexibility and quality.

Strategy and production performance

(15)

15 specific business strategy”. In manufacturing a product is produced that should be beneficial for the customer which purchases the product. Treacy & Wiersema (1993) define this as delivering value to the customer. Originally customers based the value of a product on a combination of the price and the quality of the product, however today other factors as for example after sales, convenience of purchase and innovativeness of the product are becoming increasingly important. To deliver value to the customer Treacy & Wiersema (1993) recognize three general strategies, customer intimacy, operational excellence and product leadership.

Customer intimacy

A company which focuses on customer intimacy is: “segmenting and targeting markets precisely and then tailoring offerings to match exactly the demands of those niches” (Treacy & Wiersema, 1993). In more recent literature however, the focus of companies pursuing customer intimacy is even more specific,: “Customer-intimate companies don't deliver what the market wants but what a specific customer wants” (Treacy & Wiersema, 1995). This specific focus requires each company to make customer specific products, which has implications for the production of these products. The production has to be flexible to be able to incorporate all the different customer specific requirements into a product. Furthermore because the aim is to provide the customer with the right product, these companies are less focused on the cost of the product. As Treacy & Wiersema, (1993) put it: “make sure the customer gets the right product, whether its retail price comes to $59 or 59 cents”. Summarizing companies focused on customer intimacy are mainly focused on flexibility and quality in order to provide customer specific products.

Operational excellence

(16)

16 predominantly focused on cost, however in order to reduce the cost and be able to deliver product in time there is also a focus on time, as well internal as external.

Product leadership

A company which focuses on product leadership is: “offering the customer leading-edge products and services that constantly enhance the customer’s use or application of the product, thereby making rivals goods obsolete” (Treacy & Wiersema, 1993). This holds that these companies should be creative and innovative and should be able to commercialize their new product quickly (Treacy & Wiersema, 1995). For this reason, not only the speed to manufacture a product is important, but also the speed in which product mix changes can be made. Other characteristics of product leadership are that companies need to “focus on the core processes of invention, product development, and market exploitation” (Treacy & Wiersema, 1995). Summarizing companies focused on product leadership will most likely measure performance as a matter of innovativeness, flexibility and time. Where flexibility is needed for quick product mix changes and time in order to bring products to the market quickly after the product mix has changed.

The previously explained value disciplines have some resembles with the generic strategies as defined by Porter (1985).The generic strategies are originally classified by a 2*2 matrix with on one axis the competitive advantage (low cost and differentiation) and on the other axis the competitive scope (broad target and narrow target). However Porter (1985) recognizes three different possible generic strategies that can be followed instead of 4 as the matrix (figure 2.1) would suggest, these are:

• Cost leadership • Differentiation • Focus

According to Porter (1985) cost leadership is when a firm focuses on becoming “the low-cost producer in de industry, which is “the clearest of the three generic strategies”. This strategy

(17)

17 closely resembles the operational excellence value discipline. With a differentiation strategy: “a firm seeks to be unique in its industry along some dimensions that are widely valued by buyers” (Porter, 1985). This strategy has some resemblance with the operational excellence value discipline, however operational excellence is focus on innovation as the differentiating attribute, where a differentiation strategy has no focus on a predefined area in which a company can differentiate itself from a competitor. Lastly the a company that pursues a focus strategy: “selects a segment or group of segments in the industry and tailors its strategy to serving them to the exclusion of others” (Porter, 1985). This strategy resembles the customer intimacy value discipline the most, however customer intimacy is even more customer specific and can be seen as an extreme form of Porter’s focus strategy.

The previously explained value disciplines and generic strategies have some similarities. However the value disciplines are more specific than the generic strategies, which makes observing differences between companies following a specific value discipline easier and better to observe. For this reason it is chosen to continue this thesis by making use of the three value disciplines of Treacy & Wiersema (1993).

2.2. Product architecture

A broadly used definition of product architecture is the definition provided by Ulrich (2005): “Product architecture is the scheme by which the function of a product is allocated to physical components”. The product architecture provides information on how many components a product has, how these components should work together and how they are build (Fixson, 2005). According to Fixson (2005) a product architecture “can be nominally defined as a comprehensive description of a bundle of product characteristics including number en type of components, and number and type of interfaces between components, and, as such, represents the fundamental structure of the product.”

(18)

18 implemented by multiple chunks. It should be noted however that products are seldom completely modular of integral.

Architecture type Modular architecture Integral architecture Implementation One chunk implements one or

a few functional elements entirely.

1. Multiple chunks are responsible for the implementation of one functional element

2. A single chunk implements multiple functional elements.

Interactions The Interactions between separate chunks are most often well defined

The interactions between chunks are often ill defined. Table 2.1: Architecture properties

* table based on the following sources: Ulrich & Eppinger, 2008, Ulrich, 2005, Robertson & Ulrich, 1998, Stone et al., 2000.

Modular architectures

One important reason for companies to use a modular architecture is because it allows for “product change” (Ulrich & Eppinger, 2008). This is more difficult for an integral architecture, because here all the components are related to multiple functions. Some of the reasons identified by Ulrich (2005) for change are:

• Upgrade – some specific functions of a product are improved.

• Add-ons – a manufacturer sells a basic product to which the user can add components. • Adaption – this is used for products which may be used in different environments. • Wear – parts of a product may deteriorate and need to be replaced.

• Consumption – some products consume materials, which can be replenished.

• Flexibility in use – modules are used to configure the product for different purposes. The list of reasons for product change is updated by Ulrich & Eppinger (2008).

• Reuse – firms may choose to change only a few functional elements of a product while retaining the rest of the product.

(19)

19

1. Slot-modular architecture - In a slot-modular architecture each interface type between components is different, and thus every component has its own specific place in the architecture of the final product. For these reasons modules with a slot-modular architecture can’t be interchanged.

2. Bus-modular architecture - In a bus-modular architecture all the components have the same type of interface. The product has a common bus to which all the different components can be connected. The different modules can be interchanged in a bus-modular architecture.

3. Sectional -modular architecture - In a sectional-modular architecture all interfaces are of the same type, just as in a bus-modular architecture. However unlike the slot-modular architecture and the bus-modular architecture the sectional-modular architecture does not contain a single element to which all the components are connected. For these reasons the modules have to coupled together to form the product.

Table 2.2 Description of modularity types * Descriptions based on Ulrich, 2005

Commonality

A topic closely related to the topic of modularity is the degree of part commonality a product possesses. The degree of part commonality is the relation between the number of components in an end-item. In this case a component is an inventory item, which could also be a raw material and the end-item is viewed as a “finished product or major subassembly that is subject to a sales forecast or customer order” (Collier, 1982).

The formula to determine the degree of commonality index is as follows:

Type of architecture

Example Why this type of architecture

Slot- modular Car radio A car radio has a slot-modular architecture because it is de-coupled from the other components of the car, and it has a different interface than other components the car possesses.

Bus- modular USB on a computer A USB port on a computer can be considered a bus-modular architecture because different components (e.g. mouse, keyboard, memory stick ) can be connected to the USB port.

Sectional-modular

A couch composed of multiple sections

(20)

20

Degree of commonality index (C) = ∑

Where Φj is the number of immediate parents component j has over a set of end items, and d is the number of distinct components in the set of end items.

Following the formula it becomes clear that with a higher commonality index the end item(s) poses more common components. The lowest level of component commonality an end-item can poses is C = 1, in this case each of the components of the end item is unique. The highest level of component commonality an end item can poses is C = d, in this case one or multiple end item(s) have only one unique component.

Product platforms

To increase the level of commonality and still maintain a certain level of variety in the product portfolio companies often make use of product platforms (Alizon et al. 2007). Although product platforms are getting more common, there are still multiple definitions in academic literature. Some of the definitions can be considered quite general while others can be considered industry specific and quite narrow.

General definitions:

1. “It is a set of subsystems and interfaces that form a common structure from which a stream of derivative products can be efficiently developed and produced” (Meyer, 1997).

2. “the collection of assets that are shared by a set of products” (Robertson & Ulrich, 1998 ), where the assets can be components, processes, knowledge or people and relationships.

3. “encompassing the design and components shared by a set of products” (Meyer & Utterback, 1993).

Narrow and industry specific definition:

1. A car platform can be described as “having a unique underbody and floor pan and serving as the foundation design for multiple models” (Macduffie et al., 1996).

2.3. Different types of facility layout

(21)

21 layouts and the delayed differentiation (hybrid) layout (Benjaafar et al., 2002). The traditional layouts are mostly designed to meet the requirements of a specific product(s) and predetermined production volumes, which are assumed to be relatively stable over a long time, at least 2 or 3 years or more. The next generation layouts however are more focused on companies which produce a multitude of products with multiple product variations, with shorter lifecycles. For this reason the new layout types need to be more flexible. This means that they should be able to cope with differing products and production volumes. To handle the fluctuations in product mix and volumes two approaches are generally suggested, these are a robust layout which stays efficient under changing circumstances or an agile layout to which frequent modifications can be made (Braglia et al., 2003).

Evaluation criteria

Historically the evaluation criteria which was mostly used to evaluate the conventional layouts is the amount of material handling, which is defined by Nicholas (2008) as the transferring of items from one operation to the next, which is non-value-added. This evaluation criteria originates from the Facility Layout Problem (FLP). FLP’s main objective is to minimize the material handling cost between machines, while arranging n resources within a shop floor, according to suitable constraints (Meller & Gau, 1996). However this evaluation method fails to capture the needs of modern companies, such as the need for flexibility. According to Benjaafar et al., (2002) for modern companies scope is more important than scale and responsiveness is more important than cost and reconfigurability is more important than efficiency.

Layout types

As explained there are many different types of facility layouts, each of these have advantages and disadvantages in differing situations. In this paragraph the different layout types are examined briefly to establish an overview of the specific characteristic of each layout type.

(22)

22

Process layout (job shop): In this type of layout similar equipment or processes are clustered together into functional area’s or departments. The end product is routed through the departments, which each perform a specific operation on the product. The general purpose of this type of layout is that each department can perform work on a variety of different products (Nicholas, 2008). Furthermore in this type of layout the products are mostly produced individually or in small batches. An advantage of this type of layout is that new products can be easily added to the production, without having to change the basic layout of the departments. For this reason this type of layout is often used in a make-to-order / engineer-to-order environment, and the customer specific product can be routed through the facility in the appropriate sequence. It should be noted however that this type of layout is subjected to a large number of possible routings and thus material handling and moving can consume a lot of time.

Product layout: In a product layout all the operations necessary to complete the production of a product are sequentially arranged in a so called production, assembly or flow line. When this type of layout is used there is usually only one or a few different products that have to be produced, which have to be produced, continuous or in a repetitive manner (Nicholas, 2008). The advantage of this type of layout is that all the necessary operations needed for a product can be placed in close proximity, which reduces the material handling and moving. A drawback of this type of layout is that it can’t handle variations in the product portfolio very well.

Cellular layout: In this type of layout a portion of the factory is focused on making a family of products or products with the same processing requirements (Benjaafar & Sheikhzadeh, 2000). This is done by grouping diverse equipment and machines and placing the manufacturing functions in close proximity to the workers (Yin, 2003). Specific advantages that are often linked to a cellular layout are that they can simplify the work flow and reduce the material handling, because they are generally designed to handle a specific set of products. However it is often mentioned that cellular factories are inefficient when there is a frequent change in the product portfolio, or when demand changes (Benjaafar et al., 2002).

(23)

23 This type of layout has as an advantage that suppliers can be changed without having to change the main layout (Benjaafar et al., 2002).

Delayed differentiation (Hybrid) layout: The delayed differentiation layout consists out of two stages and sometimes it is a combination (hybrid) between two types of layouts. The first stage of this layout type makes undifferentiated products, and resembles a product layout. In this stage the intermediate products are produced in a make-to-stock fashion. In the second stage the products are customized to fit the needs of specific customers. This second stage can be in the form of a product layout, but can also be in the form of a process layout or even a cellular layout. The production in the second stage is usually based on the actual demand and thus products are produced in a make-to-order / assemble-to-order fashion.

Partially distributed layout: In a partially distributed layout similar equipment or processes are clustered together just as in a process layout, however a cluster of similar equipment or processes can be found in multiple distributed locations throughout the factory. With this type of layout duplicate departments are placed strategically throughout the factory, to be able to better withstand fluctuations in job-flow patterns (different products) and volumes than a process layout (Benjaafar et al., 2002).

Maximally distributed layout: The maximally distributed layout resembles the partially distributed layout and is also created to better withstand fluctuations in job-flow patterns and volumes. However the difference between the maximally distributed layout and the partially distributed layout is that in a maximally distributed layout there are no clusters of processes or similar equipment, because each type of equipment is placed individually throughout the factory.

(24)

(25)

25

3. Conceptual model and hypotheses

Based on the literature review a model showing the discussed elements is constructed (figure 3.1). The relationship between the company focus and the corresponding production performance measures is discusses first and can be seen in detail in figure 3.2. The remaining relationships, as shown in the model, are explained in next sections followed by the construction of three hypotheses. Each hypothesis is focused on a specific value discipline. The arrows in the model indicate that there is an expected relation between the connected elements. The arrows do not indicate positive or negative influences because there is no better or worse, because each element contains types (e.g. type of product architecture or type facility layout) and doesn’t contain sliding scales which can be influenced positively or negatively.

3.1. Relationships between company focus and production

performance measures

Combining the different value disciplines with the literature about the most common production performance measures a model which shows the expected relationships is constructed (figure 3.2). The relationships as shown in the figure are the performance measures that are expected to be the chosen by companies as the dominant measures, given the choice for a specific value discipline. A company focused on operational excellence is expected to have a strong focus on costs in order to ensure a competitive price of the product. Next to this these companies are also expected to have focus on time in order continuously try to eliminate production steps to reduce transaction and other costs (Treacy & Wiersema, 1993). A company focused on customer intimacy is expected to focus less on cost and more on flexibility in order to be able to deliver the product that a specific customer wants. Furthermore the customer is expected to get a quality product which fits his or her needs, for this reason a company focused on customer intimacy is also expected to have a strong focus

(26)

(27)

Identifying relationships between the company focus, facility layout and product architecture 8/14/2011 Klaas Werkhoven S1830740 27 Performance Measures Company focus Customer intimacy Operational excellence Product leadership Costs • Material cost • Labour cost • Inventory costs • WIP Levels • Cost of quality Non costs

Quality: Produced quality

Perceived quality

Customer satisfaction (e.g. mismatch error) Rework

Time :

Flexibility: Ability to make rapid: Design changes

Volume changes (Total volume) Product mix changes

Volume changes in batch size Ability to offer: Large degree of different products

Internal: Run times Setup times Wait times Move times External:

Supplying lead times Manufacturing lead times Distribution lead times Delivery speed

Reliability of delivery speed Time to market

Environmental: Environmental friendliness of products

Environmental friendliness of production process Prevention of environmental incidents

Perceived environmental friendliness (image)

Innovation: Innovative products

Product high in R&D content

(28)

28

3.2. Relationships between production performance measures and

product architecture

As explained in the literature review it is necessary to produce products that are beneficial for the customer, value should be delivered to the customer. According to Ulrich & Ellison (1999) even if different products use the same technology, different approaches for the product development are necessary if the customer requirements are different. For the designing phase of a product this holds that a company should focus on those components that “map to multiple customer requirements in difficult-to-articulate ways“ (Ulrich & Ellison, 1999).

Furthermore according to Ulrich & Ellison (1999) in order to achieve maximum performance, it is beneficial to focus on those components that are used to bind the other parts together, in terms of functionality and interfaces. It should be noted however that this is only possible when a function is implemented by one specific component or chunk. If a product has an integral architecture it becomes much more difficult to focus on a specific part or chunk. The component used to bind the parts together can be seen as a product platform which can have several benefits / implications for the production performance. According to Muffatto (1999) there are three important performance drivers for product platforms, these are: cost reduction, productivity of development and development lead time reduction. Costs can be reduced because “parts and assembly processes developed and tested for one model do not have to be developed and tested for the other” (Robertson & Ulrich, 1998). Furthermore there can be reductions in capital investment (Muffatto, 1999). The productivity of the product development can be enhanced, because “platforms allow both the number of prototypes and resources devoted to testing to be reduced” (Muffatto, 1999). Lastly platforms can reduce the development lead time if the platform can be reused across product families (Fixson, 2005). Because of the advantages of product platforms: “companies are using platform-bases product development to create product families that provide sufficient variety for the market while maintaining economies of scale and scope within their manufacturing process” (Simpson, 2004).

(29)

29 way that the degree of commonality is high, allows a company to increase its performance in multiple ways. First by increasing the degree of commonality costs can be reduced by scale effects. If a common component is needed in large quantities this might provide opportunities to lower the production cost. In the case that the components are bought from a supplier, there might be a reduction the purchasing price. Next to this transaction cost can be lowered. Second by increasing the degree of commonality inventory holding cost can be reduced, while at the same time keeping service levels constant (Collier, 1982). The holding cost can be reduced because the cumulative required safety stock for a component used in multiple products is lower than when safety stock is held separately for each product. Another choice companies can make is to keep the safety stock at the level which is required if safety stock is held separately for each product. In this case the costs stay the same, however the service level can be improved (Collier, 1982), which could for example increase the reliability of the delivery speed.

The degree of modularity can have several implications for the performance of the production. It has been claimed that products are becoming more modular, however according to Fixson (2005): “there are multiple forces working simultaneously” and this doesn’t necessarily have to lead to a more modular architecture in order to increase performance. Again this has to do with the multidimensionality of performance. If for example flexibility (e.g. the ability to offer a large degree of product variations) is viewed as the main driver of performance, than there are arguments in favor for modular architectures. This is because in the most extreme case of modularity each functional element can be replaced by changing the responsible component. When using a fully integral architecture all the components have to be changed in order to change one functional element.

(30)

30 Next to the previous mentioned arguments in favor for modular architectures, modular architectures also make it more likely that a component will be commonly useful (Ulrich 2005). Another performance measures that could be used by companies is for example quality (e.g. produced quality, perceived quality or customer satisfaction). Customer satisfaction can for example be based upon the degree to which the product delivers the performance the customer requires. These performance characteristics are dived in two categories by Ulrich (2005), local performance characteristics and global performance characteristics. Local performance characteristics are “those performance characteristics that arise only from the physical properties of a local region of the product” (Ulrich 2005). Global performance characteristics are those characteristics that arise “from the physical properties of most, if not all, of the components of the product” (Ulrich 2005).

The local performance characteristics can be optimized when a modular architecture and when an integral architecture is used. However “global performance characteristics can only be optimized through an integral architecture” (Ulrich 2005). Because of this it seems logical that especially in the cases where customer satisfaction is related to global performance characteristics integral architectures are used. An example of this is if a car manufacturer wants “to achieve a particular noise/vibration/harshness level in cars at different maximum speeds, engineers need a deep understanding of the subtle linkage between the body, chassis, engine, and drive-train” (Eggen, 2003). It should be mentioned however that there are some subtleties in this case. This is because what might be considered a component by one company might be the end product of another company. In this way a product with an integral architecture might be used as a module in a larger highly modular product.

Lastly in some cases integral architectures are also used to reduce the overall production cost. This can be done for products which have to be produced in high volumes and when raw material cost contribute significantly in the overall costs (e.g. pens or scissors).

(31)

31 characteristics and aiming a specific customer requirements integral architectures seem more suitable.

3.3. Relationships between production performance measures and

facility layout

Several authors mention that capturing the relationships between operational performance and a layout configuration is notoriously difficult, and almost no papers have been written on this topic (Benjaafar et al., 2002 & Yang, 2000). Furthermore as explained in the literature review historically facility layouts are evaluated on the amount of material handling. However according to Drira et al., (2007) the choice for a particular layout design depends on the product variety and production volumes. Benjaafar et al., (2002) and Hegde et al., (2005) add flexibility, in the form of scope and reconfigurability, to the list of criteria which determine the appropriateness of a certain layout. The different facility layouts as discussed during the literature review are linked to one or more performance dimensions for which they are assumed or claimed to be beneficial.

The fixed position layout is assumed to be beneficial, in terms of cost, when the product that has to be produced is very large, which could make it expensive to move (Nicholas, 2008). Furthermore it can be argued that products which are very large are often produced to special customer requirements which could make it too expensive to have specialized stations because much of the work is only done once and is customer specific.

The process layout can be seen as the conventional layout that is well equipped for flexibility. This layout type can accommodate changes in products and some changes in volumes, without having to change the layout, which could be costly. However it should be noted that because of the arbitrary way of dealing with variety costs, especially handling cost, can grow exponentially when sequential operations aren’t in close proximity of each other (Nicholas, 2008).

(32)

32 line (Nicholas, 2008). Disadvantages are that this type of layout is mostly focused on the production of one kind of product, which makes it rather inflexible.

A cellular layout can, in terms of performance, be seen as a hybrid between a product layout and a process layout. According to Nicholas (2008): “a workcell is actually a dedicated job shop with features of a repetitive product layout”. Although a cellular layout can reduce material handling and accommodate some variety it doesn’t excel at either of those (Benjaafar et al., 2002).

The spine layout resembles a product layout however smaller assembly lines are attached to the main assembly line. This type of layout is also build for low levels of material handling cost. However this type of layout is focused on producing products with a modular layout (Benjaafar et al., 2002). The modules that are added to the product can be from internal as well as external suppliers (Ulrich, 2005). By making use of these product characteristics the spine layout can be cost efficient while at the same time enabling companies to be more flexible, because the can easily produce multiple product variations.

A delayed differentiation layout also makes use of the modularity of the produced product to be cost efficient and at the same time maintaining some degree of flexibility. The first stage is mainly designed to be cost efficient, while the second stage provides the opportunities to customize the product with job-shop like components (Benjaafar et al., 2002).

The partially and maximally distributed layout provide flexibility and try to reduce the overall cost spread over a longer period of time, where the demand fluctuates and where new or different products are introduced regularly. This can be done because different processes are placed in such a way that there is always a routing possible were the material handling cost aren’t disproportionally high (Benjaafar et al., 2002). These routings can be seen as (flexible) temporary work cells which can be created quickly throughout the factory.

(33)

33 flows specific to the facility” (Irani & Heng, 2000). Thus by making use of modules it becomes easier to change small parts of the facility to suit the process requirements of a specific product.

Based on the descriptions of the relationships between performance measures and facility layouts, it becomes clear that conventional layouts are much more focused on cost reductions, given specific product and production characteristics. On the other hand the “next generation facility layouts” are more focused in providing flexibility either by making the layout easier to change, making the layout persistent against fluctuations or by making use of product characteristics (e.g. modularity) to enable some degree of flexibility. It is striking however that no literature could be found on how the various layouts compare to each other in terms of quality of the produced products. Furthermore no specific literature could be found on how well the various layouts could handle innovation in the products, despite the fact that this is tightly related to the concept of flexibility.

3.4. Relationships between product architecture and facility layout

This sub-chapter is divided into two parts, first will the direct relationships between product architectures and facility layouts be explained followed by a discussion about their shared ability to create a certain market response.

Direct relations

To some extent some of the relationships between product architecture and facility layout already surfaced during the discussion on the relationships between production performance and the facility layout. Especially the new types of layout (e.g. spine and delayed differentiation layout) make use of the modularity of products to increase the flexibility of the production process, without dramatically increasing the production costs.

Next to previous mentioned example other examples are available on how decisions regarding a product architecture interact with facility layout issues. Ulrich & Eppinger (2008) recommend a four step method for the establishment of a product architecture. These steps are:

• Create a schematic of the product. • Cluster the elements of the schematic. • Create a rough geometric layout.

(34)

34 Especially during the second step of this method will the product architecture be influenced by the facility layout. According to Ulrich & Eppinger (2008) functional elements should be clustered together when they use the same production technology, for more economical production. Furthermore according to Ulrich & Eppinger (2008) modules should be made for vendors which have specific capabilities. This seems especially true when a company makes use of a spine layout. In both the previous examples the degree of modularity is raised to reduce production costs and in the second example the quality is also raised by making use of specialized vendor capabilities.

Erixon (1996) goes even further and states that in order to create an “ideal ” factory, “re-engineering of the product and de factory has to proceed concurrently”, furthermore he states that “when a product is modularized, the factory should be reorganized accordingly”. To make use of already developed capabilities Erixon (1996) suggests that companies should carry over a part of a sub-system or product from an earlier generation so it can be reused. When such a “carry over” strategy is used it results in a longer life-cycle which in turn could justify larger investments (E.g building a special assembly or flow line or work cell).

When examining the rations between a product architecture and the facility layout, it should be noted that there are almost no direct relationships between a specific layout type and a specific architecture. However it can be stated that for a delayed differentiation and a spine layout the product should at least have some degree of modularity and that the product should contain a product platform. Furthermore a large size or weight of a product seems a requirement to justify a fixed position layout. However despite these specific properties of the product this isn’t directly related to the architecture of the product.

As explained in the previous chapters a company can to deliver value to the customer in multiple ways. In doing so a company selects performance measures which also influence the decisions made regarding the product architecture and the facility layout. However given the choices made by a company, the company can choose the respond to the market in various ways. The way in which a company decides to deliver value to the customer affects the way in which a firm can best respond to the market. The different options that are identified by examining literature are:

(35)

35 • make to order.

• engineer to order.

How well a company can respond to the market in one of the previous mentioned ways depends upon which product architecture of facility layout is used by the company. If for example a company uses a product layout it is easier to make products to stock. However when using a product layout it could be far from optimal to engineer products to order, because this would require to a change in the layout for every new product produced. Because of this the four identified ways in which a firm can best respond to the market, are examined in further detail and are included in this research.

Firm response to the market

In a make to stock situation every item is produced to stock, and can be delivered from stock once an order arrives. Make to stock products are normally products with a higher demand (Rajagopalan, 2002), next to this, these product can only be produced to stock if they are not customer specific. A make to stock strategy seems to suit a product layout best since both are specialized in dealing with higher demands. In addition a make to stock strategy does not seem related to a specific product architecture. An advantages of a make to stock strategy is that a uniform production schedule can be made (Nicholas, 2008), which in turn could make planning (e.g maintenance or work force) or purchasing of raw materials easier. Another advantage is that the product can be delivered directly to the customer when needed. However there are storage costs for storing the finished products.

An assemble to order strategy holds that components are stocked and assembled once an order arrives (Hillier, 2000). Products produced according to an assemble to order strategy should have at least some degree of modularization. The different (common) standardized parts of the products can be assembled in a number of different options (Amaro et al., 1999). This type of strategy seems especially suitable to be produced in a factory with a delayed differentiation layout or spine layout. It should be noted however that in a delayed differentiation layout the base product or product platform which is produced in the first stage is often made in a make to stock fashion. An advantage of this type of strategy is that by “introducing common component that replaces a number of unique components reduces the overall level of safety stock required to meet service level requirements” (Hillier, 2000).

(36)

36 (Amaro et al., 1999). Furthermore according to Amaro et al., (1999) make to order products can be standardized or can have some degree of standardized customization. Next to this products are usually made to order by companies which offer a high degree of varieties in low volumes. A make to order strategy seems applicable for every type of architecture. However because of the large degree of variety and low volumes it seems likely that a process, partly distributed or maximally distributed layout is used for the production of these items. An advantage of this type of response to the market is that inventory cost can be held low despite the variety and low volumes of products produced. The disadvantage however is that the delivery speed is lower than when a make to stock strategy is used.

In an engineer to order environment “Products are manufactured to meet a specific customer's needs and so require unique engineering design or significant customization. Each customer order results in a unique set of part numbers, bill of material, and routing”(Amaro et al., 1999). Because of the uniqueness of each item produced this strategy is not suitable in combination with a product layout. Next to this the combination with a modular layout seems illogical because modules are mostly designed based upon reoccurring operation sequences which are less likely to occur when the products produced are customer specific. Facility layouts that are more likely to be used, because they can handle the variety better are the process layout, fixed position layout, partially distributed layout and the maximally distributed layout. Furthermore because each product is customer specific it seems less likely that these product have a high degree of modularization.

3.5. Hypotheses

Three hypotheses are constructed based upon the explained relationships between the company focus, performance measures, product architecture, facility layout and the firm response to the market. Each hypothesis is focused on a specific value discipline.

Customer intimacy

(37)

37

Operational excellence

Hypothesis 2: A company focused on operational excellence will most likely chose dominant performance measures in the categories cost and time. This in turn will most likely lead to an average to high degree of modularity for the product architecture and can best be made under a product layout, spine layout or delayed differentiation layout. The firm response to the market will most likely be make to stock (figure 3.4).

Product leadership

Hypothesis 3: A company focused on product leadership, will most likely chose dominant performance measures in the categories time, flexibility and innovation. This in turn will most likely lead to an average to high degree of modularity for the product architecture and can best be made under a, spine layout, cellular layout, delayed differentiation layout or modular Company focus: Customer intimacy Performance measures: Quality Flexibility Product architecture: Integral / low degree of modularity

Facility layout: Process layout Fixed position layout Partially distributed layout Maximal distributed layout

Firm response to the market: Assemble to order Engineer to order Make to order

Figure 3.3: Conceptual model hypothesis 1: Customer intimacy

Company focus: Operational excellence: Performance measures: Cost Time Product architecture: Average to high degree of modularity Facility layout: Product layout, Spine layout

Delayed differentiation layout

Firm response to the market: Make to stock

(38)

38 layout. The firm response to the market will most likely be make to stock or assemble to order(figure 3.5). Product architecture: Average to high degree of modularity modularity Company focus: Product leadership Performance measures: Time Flexibility

Innovation Facility layout: Spine layout Cellular layout,

Delayed differentiation layout Modular layout

Firm response to the market: Make to stock Assemble to order

(39)

39

4. Research design

In order to verify the constructed hypotheses the choice is made to make use of case studies. One of the reasons for using case studies is because a case studies can investigate “a contemporary phenomenon within its real-life context, especially when the boundaries between phenomenon and context are not clearly evident” Yin (2003B). This seems to be the case in this research, because although three clear hypotheses are constructed, it could well be that there are forces involved, that are not explored in the literature review, that contribute to the decisions made by companies.Furthermore in an ideal situation the relationships between the variables used in this research would be examined with statistics and a large enough sample size (n) to create an empirical model. However it seems impossible to create a large enough sample size within the timeframe of this research project. Next to this, the relationships which are examined in this research project are ill defined in current literature. For these reasons it seems impossible to construct a tool which can deliver unambiguous data.

4.1. Research method

The method of choice for this research is the six step method of soy (1997) which draws upon work of several case study researchers.

1. Determine and define the research questions.

2. Select the cases and determine data gathering and analysis techniques. 3. Prepare to collect the data.

4. Collect data in the field. 5. Evaluate and analyze the data. 6. Prepare the report.

The first step is discussed in chapter 1. Appendix I shows considerations that are taken into account during this step. The steps four and five are discussed in chapter 5: analysis and results. The steps two and three are discussed in the following sections.