Modular decomposition logics and the perceived benefits of applying modularity in product-service systems

(1)

Modular decomposition logics and the perceived benefits of

applying modularity in product-service systems

Master thesis, MSc Supply Chain Management

University of Groningen, Faculty of Economics and Business

June 22, 2015

By

Mark Willems

Student number: S2344351

Email: m.r.willems@student.rug.nl

Supervisor: dr. H. (Manda) Broekhuis

Second assessor: dr. S.A. (Carolien) de Blok

Ackknowledgements:

I would like to thank dr. Manda Broekhuis for her on-going support and helpful feedback throughout the process of writing this Master’s thesis. Furthermore, I would like to thank all the interviewees for their

(2)

2

Abstract

This Master’s thesis examines how modular decomposition logics can be applied in product-service systems, and what the perceived benefits are of applying modularity in product-service systems according to servitized manufacturing firms. The purpose of this study is to find out how benefits of a modular design can be applied in product-service systems in order to find a right balance between customization and costs of a product-service system. The empirical part of this research paper consists of exploratory interviews with seven servitized manufacturing firms. The following decomposition logics were identified: single level outcome oriented, single level process oriented, multilevel outcome

oriented, multilevel process oriented, and multilevel combined orientation. The modularity types that followed were mix-, cut-to-fit-, bus-, and sectional modularity. Furthermore it turned out that

customization is not a benefit of service modularity in product-service systems, because customization in the B2B market can only be spoken of when the product-service combination is genuinely tailor-made. This study has tested the usability of modularity concepts in product-service systems, which is useful for researchers and practitioners, because it contributes to a common terminology and method for modular service decomposition.

(3)

3

1 Introduction

Today, an increasing amount of manufacturing firms offer a combination of products and services to their customers (Neely, 2007). These are so called ‘product-service systems’ (PSSs). Often traditional manufacturing firms were not able to rely solely on products as value-adding output to stay competitive anymore (Yu et al., 2008), and therefore nowadays they offer a service package in combination with their products. PSSs matter because they yield economic gain, might reduce environmental impact due to providing services such as the recycling or refurbishing of their old products (Aurich et al., 2006; Baines et al., 2007), and might increase the usability of their products (Finne et al., 2013). However, the transition from manufacturing to such a servitized manufacturing firm also creates big challenges in keeping costs down and in delivering the larger range of products and services in an efficient way (Gebauer et al., 2005; Neely, 2009). Visnjic and Van Looy (2013) report in their findings that many servitized manufacturing firms gain from services on the short-term, however the service paradox (Gebauer et al., 2005) makes that the long-term profitability decreases. This service paradox suggests that servitized manufacturing firms generate higher revenues, but they tend to generate lower net profits as a percentage of these revenues (Neely, 2009). One of the actions servitized manufacturing firms can take is the application of economies of scale to their services to achieve long-term profitability (Visnjic and Van Looy, 2013). This is where the concept of modularity comes in. Modularity might add to this efficiency aim as it enables companies to provide a high variety of products and/or services at relatively low costs (Starr, 1965; Muffato, 1999; Hsuan and Skjott-Larsen, 2004; Voss and Hsuan, 2009; De Blok et al., 2014). One important question remains, which is how these benefits of a modular design can be applied in product-service systems in order to find a balance between customization and costs.

(6)

6 and Böhman, 2013; Rajahonka, 2013; Dörbecker et al., 2015; Carlborg and Kindstrom, 2014; Van der Laan et al., 2015). To the knowledge of the author, so far only Shen et al. (2012) have studied modularity in product-service systems. In their paper, they propose a product-service configuration system based on, among others, modularity principles.

When servitized manufacturing firms plan on shifting towards a modular system design, one of the actions they should take is the decomposition of their product-service offering. A decomposition logic (Van der Laan et al., 2015) should be used for this, because a decomposition logic helps defining the specific functions of modules. This paper aims to increase insight in how modular decomposition logics can be applied to product-service systems, what the considerations are for decomposing a PSS in a certain way, how to connect the modules again, and what the potential benefits could be of applying modularity to product-service systems. This should provide valuable insights to manufacturing firms who have the ambition to servitize, or servitized manufacturing firms who struggle with the challenges of servitization. Furthermore, Carillo et al. (2015) justify the importance of this paper by mentioning in their further research that the application of modularity principles to integrated product-service systems should be studied. Also the perceived benefits of modularity in PSSs (according to servitized

manufacturing firms) are studied, to test if the perceived benefits for service modularity (Rajahonka, 2013) also hold for PSS modularity. It is a valuable contribution to modularity research to add the views on modularity of PSS-providers and in particular the usability of the typology proposed by Van der Laan et al. (2015) to product-service systems. Finally, paper also tries to identify certain types of modularity that are used in product-service systems in practice. By doing this, a contribution is made to the study by Van der Laan et al. (2015) who propose a typology of six types of modularity based on literature

research. These modular types have not yet been tested for usability in practice and therefore it will be relevant to study how applicable they are in product-service systems. The research question of this paper is the following:

How can modular decomposition logics be applied in product-service systems, and what are the perceived benefits of applying modularity in product-service systems according to servitized manufacturing firms?

(7)

7

1. What is the basic choice in decomposing product-service systems? 2. Which decomposition logics can be recognized in PSS offerings?

a. What is the decomposition orientation of PSS-providers?

b. How many decomposition levels do PSS-providers apply to decompose their product-service combinations?

3. Which modularity types can be recognized in product-service systems?

4. According to the views of PSS providers, what are the benefits of applying modularity to their PSS offerings?

(8)

8

2 Research framework

2.1 Product-service systems

Product-service systems (PSSs) are a combination of servitization of products and productization of services (Morelli, 2003; Beuren et al., 2013) and they make up the future for production companies in order to stay competitive. Goedkoop et al. (1999) define a production-service system as a system which consists of an offering of products and services which provides functionality for consumers, while reducing environmental impact (Beuren et al., 2013). Examples of services that are delivered next to a product offering are product-maintenance, financing schemes, supply of consumables, take-back agreements, advice and consultancy and product-lease, -renting, -sharing, pooling (Tukker, 2004).

PSS systems may be viewed from the following perspectives (Tukker, 2004), namely,

(1)product-oriented, which means the supply of products with additional services, (2)use-(1)product-oriented, meaning that the

use of the product is sold together with the services that add value to it; the product remains property of company that sells it and customer pays for the use, and (3)results-oriented, which means that only a result or competency is sold to the customer; the product remains property of company that sells it, and the customer pays for result of using it, like washing clothes instead of buying a washing machine. These perspectives are depicted in figure 1.

(9)

9 Benefits of PSSs can be identified from four perspectives (Beuren et al., 2013), namely for consumers, for providers, for the environment and for society. Benefits for the consumer are among others flexible and customized services, which leads to a higher satisfaction of the delivered product (Aurich et al., 2006). The benefits for providers of PSSs are customer loyalty (Aurich et al., 2010), the reduction of costs and resources (Mittermeyer et al., 2010), and the potential for innovation, because by offering a PSS, firms can differentiate from one another (Tukker and Tischner, 2006). The environmental benefits are due to the fact that the provider takes responsibility for the products and services by recycling and refurbishing old products (Baines et al., 2007), and finally the benefits for the society are the increase of the supply of services which creates new jobs (Baines et al., 2007; Gao et al., 2009).

Though the implementation of services in production systems seems to bring only benefits, PSS

providers struggle with the implementation of services into their production systems. This is because the implementation of services comes with high costs and investments, but they do not necessarily generate the expected higher returns. Employment costs in servitized firms are higher than in pure manufacturing firms. Furthermore the working capital per employee in servitized firms is higher and the total assets per employee are also higher. The reasons for this are likely that staff of complex services like consultancy is likely to be more expensive than staff with a limited set of skills (Neely, 2009).This is phenomenon is called the ‘service paradox’ (Gebauer et al., 2005; Neely, 2009). According to Neely a considerable amount of servitized firms go bankrupt, and therefore the transition from products to a product-service combination can be considered to bring challenges. These challenges consist mainly of higher cost and lower revenues. Gebauer et al. (2005) propose actions on how to overcome this service paradox, including establishing a market-oriented service development process, defining a clear service strategy, establishing a separate service organization, and creating a service culture. Visnjic and Van Looy (2013) add to this that product-service firms should apply economies of scale to their services in order to stay profitable on the long run. One way of achieving economies of scale is the application of modularity to service processes.

2.2 Modularity

(10)

10 definition, a product or service can be highly modular, or the opposite, which is highly integrated

(Brusoni and Prencipe, 2001; Mikkola and Skjott-Larsen, 2004; Salvador et al., 2002; Ernst, 2005; Rajahonka, 2013). Definitions of what a module exactly is varies in literature, but one of the most frequently used definitions is that a module is strongly connected internally, but relatively weakly connected to other modules (Baldwin and Clark, 2000; Schilling and Steensma, 2001; Ernst, 2005; Arnheiter and Harren, 2005, 2006; Lau et al., 2007; Bask et al., 2010). Another clear definition is one from Salvador (2007) who states that a module is the smallest building block of which a system consists. Modularity is achieved by standardizing bits of a product or service, which in turn can be combined with other standardized bits of that product or service. Rajahonka (2013, p. 47) defines a service module as: “a relatively independent part of a system with a speciﬁc function and standardized interfaces, where the system can be, for example, a service, a service production process or an organization or a network of organizations”. In this paper we define a module as a building block of a system, which can be a service, a process, a product, or a product-service combination that is strongly linked internally, but relatively weakly linked to other modules, having a specific function and highly standardized interfaces (Baldwin and Clark, 2000; Schilling and Steensma, 2001; Ernst, 2005; Arnheiter and Harren, 2005, 2006; Lau et al., 2007; Salvador, 2007; Bask et al., 2010; Rajahonka, 2013).

By making products or services modular, companies succeed in organizing complexity in an efficient way, which enables customization, product postponement and outsourcing (Baldwin and Clark, 1997; Voss and Hsuan, 2009). Other benefits from this approach are that a large number of customers can be satisfied by offering a limited degree of customization against low costs (Starr, 1965; Muffato, 1999; Hsuan and Skjott-Larsen, 2004; Voss and Hsuan, 2009; De Blok et al., 2014), flexibility (Lau et al., 2007; Pekkarinen and Ulkuniemi, 2008), cost savings (Jose and Tollenaere, 2005; Pekkarinen and Ulkuniemi, 2008; Rahikka et al., 2011), and economies of scale (Baldwin and Clark, 2000; Hölttä-Otto, 2005). In earlier literature, modularity has mostly been studied as product, process, and organizational modularity (Salvador, 2007). As from 2008, there has come an increased interest for the concept of service

(11)

11 applicable to modularity in services or products, but it is not clear if these benefits are the same for product-service systems. This leads to the following sub-research question:

According to the views of PSS providers, what are the benefits of applying modularity to their PSS offerings?

Schilling (2000) and Ulrich (1995) argue that every modular system is based on three core principles, which they call modular design principles. These imply that every module should (1)have a specific

function, (2)be a relatively independent part of a system, and (3)have a standardized interface. To be

able to achieve these modular design principles in a product-service system, one should first look determine if the product-service system should be decomposed as combinations of the product(s) with service modules, or if the products and services should be decomposed separately from each other. The next step is looking at the decomposition logics proposed by Van der Laan et al., 2015, because decomposition logics help in decomposing the product-service combinations in functional elements by defining the specific functions of modules. This concept is discussed in the next sections of this paper.

2.2.1 Decomposition logics and modular types

When product-service firms are trying to make the transformation to a modular system design, the obstacle that they need to overcome is to decompose their product-service offering in smaller parts, i.e. modules. Modules have to be defined, boundaries between them have to be established, there has to be some idea about the degree of coupling between the modules, and the interfaces between modules have to be clear (Van der Laan et al., 2015). But more general, a decomposition logic helps defining the specific functions of modules which is one of the modular design principles (Ulrich, 1995; Schilling, 2000). Van der Laan et al. (2015) propose a decomposition logic to answer these questions. They define a decomposition logic as the explanation of how a service offering is decomposed into modules (Van der Laan et al., 2015). Furthermore, van der Laan et al. (2015) distinguish three consecutive (and partly iterative) modular design steps from theory, namely (1)determining the decomposition orientation,

(2)determining levels of decomposition, and (3)the identification of the most relevant interdependencies and separating them (Brusoni, 2005; Simon, 1962; Ulrich, 1995).

(12)

process-12 orientation (‘how’ it is delivered). A design choice has to be made, which leads to a process-orientation, outcome-orientation, or combined orientation. Outcome oriented modules are operationalized as that not all modules have to be delivered, the modules can be delivered separately from each other, and there is no predefined sequence in which the modules have to be delivered. Process modules however, follow a predefined sequence of process steps which can only be adapted in time and length.

This leads to the following sub-research questions:

Which decomposition logics can be recognized in PSS offerings? What is the decomposition orientation of PSS-providers?

The second step in the decomposition typology proposed by Van der Laan et al. (2015) is the design choice of whether the system needs to be decomposed on an overall service offering, for example decomposition of post and packaging services into letter, e-post, package, and warehousing, or on a detailed level of activities (e.g. the activities within these modules, such as printing, picking up, and sorting) (Bask et al., 2010; Van der Laan et al., 2015). A decision for the first choice leads to

decomposition on a single-level, and a decision for the second choice leads to a decomposition on multiple levels. This leads to the following sub-research question:

How many decomposition levels do PSS-providers apply to decompose their product-service combinations?

(13)

13

Service offering is decomposed into multiple outcomes or themes on a single

decomposition level, characterized by pooled dependencies

Service offering is decomposed into multiple phases on a single decomposition

level, characterized by sequential dependencies

Service offering is decomposed into multiple phases on multiple decomposition levels, characterized by

sequential dependencies Service offering is decomposed into

multiple outcomes or themes on multiple decomposition levels, characterized by

pooled dependencies Outcome-oriented Process-oriented Si n gl e le ve l M u lt ile ve l

Service offering is decomposed into multiple phases and outcomes on a single

decomposition level, characterized by pooled and sequential dependencies

Service offering is decomposed into multiple phases and/or outcomes on multiple decomposition levels, characterized by pooled and sequential

dependencies

Combined orientation

Figure 2 – Overview of decomposition logics separate service modules (source: Van der Laan et al., 2015)

(14)

14 Outcome-oriented Process-oriented Combined orientation Single level combined

PSS modules

Service offering is

combined with product

offering and decomposed into multiple outcomes or themes on a single decomposition level, characterized by pooled dependencies Service offering is

offering and decomposed into multiple phases on a single decomposition level, characterized by sequential dependencies Service offering is

offering and decomposed into multiple phases and

outcomes on a single decomposition level, characterized by pooled and sequential dependencies

Multilevel combined PSS modules

Service offering is

offering and decomposed into multiple outcomes or themes on multiple decomposition levels, characterized by pooled dependencies Service offering is

offering and decomposed into multiple phases on multiple decomposition levels, characterized by sequential dependencies Service offering is

offering and decomposed into multiple phases and/or

outcomes on multiple decomposition levels, characterized by pooled and sequential dependencies

Table 1 – Decomposition logics product-service modules

The decomposition of a product-service system can also result in different types of modules. To characterize these modules we refer to modularity types as proposed by Ulrich and Tung (1991). These are further modified by Van der Laan et al. (2015) into the following service modularity types:

component-sharing modularity, component-swapping modularity, mix modularity, cut-to-fit modularity, bus modularity, and sectional modularity (figure 3). The decomposition of any offering should lead to

one or multiple of these modularity types. A relation has been found between the chosen

(15)

15

Time

Component sharing: Reuse of standardized

modules. Outcome, process and interface(s) are standardized.

Mix: Offer pre-defined choices with regard to

process aspects, thereby influencing the overall service experience. Process and interface(s) are standardized.

Bus: Adding or subtracting modules to or from a

basic service offering. The basis service and interfaces are standardized.

Sectional: Unrestricted combination of modules

into a service offering. Interfaces are standardized.

Source: adapted from Ulrich and Tung 1991

Cut-to-Fit: Offer options to adapt process aspects

of a service module during delivery without affecting other modules. Outcome and interface(s) are standardized.

Component swapping: Offer mutually exclusive

choice options within a fixed service offering. Outcome, process and interface(s) are standardized.

Legend

Inner circle, box, triangle: service outcome dimension Outer circle, box, triangle: service process dimension Black: standardized

Time

Figure 3 – Overview of modularity types (source: Van der Laan et al., 2015)

The relationship between the decomposition logics and the modularity types are depicted in table 2 (Van der Laan et al., 2015).

Table 2 – Relation between decomposition logic and modular types (Van der Laan et al., 2015)

The relevant literature for the empirical part in this study was structured as follows. First definitions and benefits of product-service systems were described, second definitions and benefits of modularity were described. After this the decomposition logics with the modularity types followed. To study the benefits of modularity in product-service systems, and to test the usability of the modular decomposition logic

Outcome-oriented Process-oriented Combined orientation

Single-level

Sectional modularity

Cut-to-fit modularity Bus modularity

Multi-level

Component-sharing modularity Component-swapping modularity

(16)

(17)

17

3 Methodology

The literature on service modularity and product-service modularity is scarce, therefore a qualitative research method is used, because this allows for the presence of the researcher in the collection of data, has a small sample size to be able to draw in-depth conclusions, is helpful for building theory, and is useful because the concept of modularity in product-service systems is not predefined (Correa, 1992).

3.1 Main method

In this study, semi-structured interviews were held to get a deeper understanding of how modular decomposition logics can be applied to product-service systems. Bask et al. (2010) indicate that in-depth interviews with servitized manufacturing firms on how modularity can be used in their offerings can create a more thorough understanding on if modularity is usable in products and services from a product-service system. Next to the semi-structured interview questions, this interview also contained several structured statements as benefits of modularity, which were found in literature. The

interviewees were asked to score these statements about the benefits on a 5 point Likert-scale. A score of 1 indicates that the interviewee strongly disagreed, and a score of 5 indicates that the interviewee strongly agrees with the statement. Also, after each statement, each interviewee was asked to give extended arguments on why he gave that particular score to the statement.

3.2 Selection of interviewees

The sample consists of seven production companies situated in the Netherlands. The companies have been selected based on the following criteria, namely that they offer products and additional services to their customers. The services that the firms offer vary, but they can be among others installation, maintenance, repair, replacement, asset management, finance, insurance and consultancy. The

(18)

18

Company

Activities

Function interviewee

Length of the

interview

Interview

mode

Interview 1 Specialist in electronic

systems and products

Project manager 43 minutes Face-to-face

Interview 2 Train manufacturer Manager service

unit

51 minutes Face-to-face

Interview 3 Truck manufacturer Team manager contract

administration

Interview 4 Specialist in electronic

systems and products

Manager supply chain and production

Interview 5 Technical automation,

electrical and mechanical engineering

Sales and project manager

Interview 6 Hygene products

industry

Supply chain coordinator

56 minutes Video conference

Interview 7 Food industry Sales manager (former

supply chain manager)

Table 3 – Interview details

Company

Products

Services

Interview 1 Electronic

systems

Industrial design, project management, process management, development and engineering, logistics and procurement, manufacturing and assembly, testing, value engineering, life cycle management, after sales.

Interview 2 Trains Preventive maintenance, repairs en overhauls,

modernization en revitalization, spare parts and subsystems, fleet management, and training

Interview 3 Trucks Workshop services, trailer services,

maintenance contracts, repair, rental, training, financial services, and spare parts

Interview 4 Radio-communication devices,

public transport systems, military defence systems

Lifecycle services, maintenance, monitoring, and operational services

Interview 5 Shore power converters Maintenance and repair

Interview 6 Hygiene products Analysis and consultancy

Interview 7 Cheese products Logistics services

(19)

19 The companies were further classified in their main manufacturing approach and whether or not they already worked with service modules (or if they have such standardized components in their services that these components could be regarded as modules). This is depicted in table 5.

Company

Manufacturing approach

PSS system

Service modules

Interview 1 Engineer to order Product-oriented Yes

Interview 2 Make to order Product-oriented No

Interview 3 Make to order Product-oriented Yes

Interview 4 Make to order/Engineer to order Product-oriented No

Interview 5 Make to order Product-oriented No

Interview 6 Make to stock Product-oriented No

Interview 7 Make to stock Product-oriented Yes

Table 5 – Interview details

3.3 Data collection

For this study, emails were sent out to potential interviewees of whom there was the expectation that their company would have a product-service system. People from the researcher’s own network, but also from his supervisor’s network were contacted with an introduction email with all the general details of this study. If the respondents were interested they could send an email back, after which the

(20)

20 using two devices, to prevent the loss of valuable data. The interviews were transcribed in Dutch, using Inqscribe software, which allows for easy switching between the interviewer and interviewee, using hotkeys. Also a timestamp is added to the transcript, which allows for a more transparent transcription, because it is easier to assess who said what and when. Transcripts of the interviews were sent to the interviewees to check if all the information is correct. The quotes that were finally used in the findings section were checked for accurate English by a proficient English speaker. The names of all the

companies and interviewees were kept anonymous, at request of most of the interviewees.

3.4 Operationalization of main variables

The operationalization of the decomposition of the product-service system, modular decomposition logics (decomposition orientation, and decomposition level), the modular types, and the benefits and disadvantages of modularity are presented in table 6. The themes from literature were transformed into interview questions, with the aim of getting a deeper understanding of the specific services, and how they should be decomposed, according to the interviewees. When for example, the aim was to find out which decomposition level the interviewees wanted to decompose their service on, an interview question was: “If modules are identified/possible: could you divide these modules into even smaller

portions?”.

(21)

21

Variable

Interview question(s)

PSS

Decomposition logics

Do you sell the services or portions of these services separately from the product?

Orientation

Would you prefer to make PSS modules as a process (police investigation example), or outcome (investment course example), or a combination of both? Why?

Level

If modules are identified/possible: could you divide these modules into even smaller portions?; Can you combine these modules easily? (i.e. module spare parts and module repair). Go more in-depth with this question; Are there any services which are indivisible? As in: they contain so much variation that it is impossible to standardize portions of them?

Modularity types Decomposition logics

Does your firm already work with modules, for example in building products? If so, in which fashion? Is there a document that clearly describes these

modules? Are there already highly standardized portions of service that you offer to your customers?; Which combinations of products and services are often chosen by customers?; Why do customers often choose these combinations?; How can the products and services exactly be combined? In time (sequential), or substantively?

Company

Benefits/disadvantages Customer

Benefits/disadvantages

Which benefit could modularity bring to a product-service system? Which disadvantages? For whom? (company and customer perspective)

Table 6 – Operationalization of main variables

3.5 Data analysis

(22)

22

4 Findings

According to Miles and Huberman (1994) each interview should be examined individually before making an overall analysis. In this section each interview will be analyzed separately from each other. First a general description is provided of the products and services of the interviewees’ company. Next, an analysis of the decomposition of the PSS, the decomposition logics, and modularity types is described. Finally the perceived benefits and disadvantages of modularity are described, which consists of two parts. The first part is the answer of the interviewee to an open interview question about the benefits and disadvantages of modularity. This is to ensure that the interviewee gives benefits with an open mind. The second part consists of structured benefits of modularity from academic literature to which the interviewees give a scale from 1 to 5 (1 strongly disagree, 5 strongly agree) and their extended view on the modularity benefits.

4.1 Interview 1 (i1) – Specialist in electronic systems and products

The first interview was conducted with a project manager from a company that is a specialist in

electronic systems and products. This company manufactures electronic systems, i.e. they engineer and produce high-tech products. The services that come along with these products are industrial design , project management , process management, development and engineering, logistics and procurement , manufacturing and assembly, testing , value engineering , life cycle management and after sales. The service which was discussed in detail was ‘design for excellence’ (DfX), which contains most of the services mentioned. The DfX program is an early stage check for risks in the development and

(23)

23

Figure 4 – Design for Excellence

Decomposition logics and modular types

The first step is to determine the decomposition of the PSS system. It is clear that the DfX program is closely linked to the product. The interviewee mentions that the design for excellence program consists of various checks, where one of the checks is life cycle management. “Life cycle management is the

monitoring of components of the product to assess in which stage of their life cycle they are at this moment, and whether or not they are already end of life”. This means that if DfX modules are

established, they are attached to the product. The interviewee confirms this: “life cycle management is a

combination that is sold together with a product”.

Regarding the orientation of the decomposition logics, DfX has a combined, i.e. outcome and process, orientation, because the interviewee mentions that all the components are “basically outcome modules,

because they can all be used separated from each other, however, design BOM scrubbing is an input for all the others. (…) but if I have to generalize I would have to say outcome modules.” This means that

although the separate modules have an outcome orientation, these modules can only be delivered in a sequential process-oriented way, i.e. after the basic module of BOM scrubbing and completion is executed. Also there are links possible between the different modules of this DfX service. For example:

“when you want to predict quality with design for quality, you need input from a specific part of design for test (…) in that sense you are making a combination between design for quality and design for test”.

Regarding the levels in the decomposition the interviewee states “(the DfX modules) exist of one, two, or

three analyses. There are multiple building blocks in this component”. This means that the DfX modules

(24)

24 Finally, the identified modularity types are bus modularity and mix modularity. The argument for the typing of bus modularity is that BOM scrubbing and completion comes first after which the other predefined outcome modules can be added. Second, the argument for mix modularity is that there are two or more components (i.e. BOM scrubbing and completion with one or more of the designs for X), which together provide a different service for the customer.

Benefits and disadvantages of modularity for the company or customer

Interviewee 1 mentions the flexibility for the company. “You do not always need to do everything that

the service has to offer”. By this he means that the services that his company provides are too extensive

to offer as a whole to the customer and therefore are in need of modular decomposition. Also, according to him modularity offers the possibility to customize the service to customer needs. Disadvantage for the company is that the development of modules consumes time and money.

4.2 Interview 2 (i2) – Train manufacturer

The second interview was conducted with the head of the service department of a train manufacturer. This company manufactures trains and their services are preventive maintenance, repairs and overhauls, modernization and revitalization, spare parts and subsystems, fleet management, and training. The discussed services in this interview are modernization, and training and documentation. Modernization takes place when the train is not fit for purpose anymore after about ten to twenty years and it needs to be revitalized. Training (and documentation) consists of a basic training in operating the train, which is always obligatory when a customer buys a train after which several options follow for more in-depth training.

The decomposition of the PSS combination is done in two ways:

(25)

25

(…) the customer does the maintenance himself on all his trains, but at some point he needs to modernize his trains and then he comes to us (…) we also do trains of other train

manufacturers”. The services are often delivered by themselves and are not strictly linked to

their products.

b. Some service modules are closely related to and an inherent part of the product. In the case of training and documentation for instance. The interviewee states: “when a customer buys a train,

it is obligatory to get training for the operation of this vehicle”, and the same counts for

documentation: “documentation is always mandatory, because otherwise you are not allowed

to operate the train”.

Regarding the orientation of the decomposition logic, training (and documentation) are modules of a combined orientation. Training modules on their own have an outcome orientation, but because you have a basic training module, to which you sequentially attach several other training modules, the overall decomposition orientation is a combined orientation. “The basic training is always the

beginning”. Modernization of a train is also always an outcome module, because there is no strict

sequence in delivering this service, and not always all services will be necessary to modernize a train. De module modernization is a single level module, because it is nearly impossible to split this service into smaller portions. “If, for instance, you want to modernize the front of the train, this is too much of a

complex task to divide this into smaller sections”. Training and documentation however, can be split into

multiple functional elements, such as basic training in operating the train. Therefore training and documentation are decomposed into multiple levels.

The modularity types that are recognized in training and documentation are mix modularity and

sectional modularity. The argument for the typing of bus modularity is that the basic training comes first after which the other predefined outcome modules can be added. Second, the argument for mix

(26)

26

Benefits and disadvantages of modularity

The benefit for the company that interviewee 2 mentions is that “in the background it saves a lot of

work. If you have to recalculate every service each time, it’s just a hassle in negotiation. Standard packages of modules saves you in sales costs (…) it’s all about efficiency for us”. As disadvantage for the

company this interviewee mentions that there are risks in pricing of the service. Not every country has the same level of pricing, and therefore you have to be very careful in offering standard prices for your modules. “If one of our training specialists goes to Azerbaijan the pricing of components, or even his

travel expenses will be much cheaper than when he does the same job here in Leeuwarden”.

4.3 Interview 3 (i3) – Truck manufacturer

The third interview was conducted with a team manager contract administration from a major truck manufacturer. The services that this truck manufacturer offers are workshop services, trailer services, maintenance, repair, rental services, training, financial services, and spare parts. The services that are discussed in this interview are the service repair and the service maintenance, but since they are often offered in combination together, these services will be treated as repair and maintenance. This is shown in figure 5. The figure also shows that the customer can choose several modules from the basic repair and maintenance module, such as engine and breaks, or gearbox, brakes and heating.

(27)

27

The basic choice of decomposition for the service repair and maintenance is that the services are

separately decomposed. The interviewee says “repair and maintenance does not have to be linked to our

own trucks. Also trucks from … (competitors) are repaired here”. This means that the decomposition of

repair and maintenance is not necessarily attached to their products, and can therefore be decomposed separately.

The decomposition orientation of repair and maintenance follows an outcome oriented decomposition, because it is not possible to identify any process steps in the service offering. Each module can be delivered separately from other modules, it does not matter which order they follow, and the customer does not have to choose all modules. See figure 5.

The module repair and maintenance is a multilevel module, because it can be divided in many smaller components, which the customer can choose of. The interviewee states the following: “depending on

the truck that you’re driving, you have the R&M (repair and maintenance) module, which consists of subcomponents like air-conditioning. One could say that R&M consists of many modules. These sub-modules have their own sub-components, but the customer cannot choose these smaller components”.

The repair and maintenance module is clearly an example of sectional modularity, because it is an unrestricted combination of modules into a standard service offering and they can be combined in many different ways.

Interviewee 3 mentions benefits for the customer, such as “less signatures from our customers”. What the interviewee means by this is that the customer does not want to make tedious decisions about small repair and maintenance details. He wants to be able to choose standardized contracts in which

everything is arranged for the specific part of the truck. “The customer wants to be unburdened, and by

means of modularity we can achieve that”.

4.4 Interview 4 (i4) – Specialist in electronic systems and products

(28)

radio-28 communication products for the military, transportation systems, and cyber security systems. The services which were discussed during the interview are repair and maintenance of public transport systems. The products that this company produces are amongst others check-in and –out machines and ticket vendor machines. The repair and maintenance for the ticket vendor machines or check-in and – out machines works as follows. When a machine stops working, the monitoring center of this firm gets a maintenance call, after which they send a mechanic who replaces the defect machine or unit with a new component or with a component which has been repaired previously. The defect unit is then being sent for repair at the repair center, or at the repair center of an external supplier. After this, the inventory is replenished with the repaired unit. See figure 6.

(29)

29

The repair and maintenance service for company 4 is delivered separated from the product. The interviewee states that they are not always the supplier of the products that they are doing

maintenance for. “At many public transport stations we do maintenance, but the machines which we

repair are not always produced by us”. This means that the service is not necessarily attached to the

product and therefore can be decomposed on a service based level.

The decomposition orientation is a process oriented decomposition, because the service consists of process steps which cannot be seen separately from each other. It all starts with a service call, and it ends with replenishing the unit to the inventory. He mentions that “those are all process steps which are

connected to other services, like spare parts”.

This all happens on a single level decomposition, because none of the process steps in the repair and maintenance module can be divided into standardized smaller portions.

The recognized modularity types are cut-to-fit and bus modularity. The argument for cut-to-fit

modularity is that all the modules (process steps) in the service can be adapted (for example in time and length) without affecting the other process steps. The argument for bus modularity is that potentially you could add or subtract modules from the service offering, for example if the unit is being sent for repair, but after checking it thoroughly, the unit does not need any repairing.

The benefits that interviewee 4 mentions are first of all for the company, namely that it “you can

(30)

30

4.5 Interview 5 (i5) – Specialist in technical automation, electrical and mechanical

engineering

The fifth interview was held with a specialist in technical automation, electrical, and mechanical engineering. Very recently they have taken over a division of a company that sells marine equipment. This division produces shore converters for luxury yachts. The interviewee is a sales manager of these products. The service that this company provides along with the shore converters is repair and maintenance. This repair and maintenance module can be divided into several smaller modules, depending on what type of servicing or repair the shore converter needs.

The basic choice in decomposing the service repair and maintenance separated from the product. The interviewee states that they do not sell the product together with the service, and that the he does not see the merits of decomposing the service together with the product.

The decomposition orientation is an outcome oriented decomposition, because not all modules are always necessary to execute, the modules can be executed separately from each other, and there is no sequence in which the modules have to be executed.

When the question rises whether or not the interviewee wants to decompose repair and maintenance into smaller portions, he states “I don’t think it has any benefits to cut the service into smaller portions,

because when you do that, you give the customer a choice where he does not want to choice. The customer has no knowledge of this complex product, and therefore does not know what to choose”. This

means that repair and maintenance is decomposed on a single level.

Repair and maintenance could take many different shapes and sizes depending on the type of repair that is necessary, therefore the modularity type that could be recognized in this service is sectional modularity.

The benefits which interviewee 5 mentions are benefits for the company, namely that “you can define

(31)

31 is increased turnover due to over specification, because the company always overcharges slightly when they offer modules to customers. These modules are standardized, and because you always want to meet customers’ demands, a module always has to be slightly more comprehensive than the customer actually needs. This also counts as disadvantage for the customer. The disadvantage for the company is according to this interviewee that it requires time and money from the company to set up a modular design.

4.6 Interview 6 (i6) – Hygiene products supplier

The sixth interview was with a supply chain coordinator from a hygiene products company. This company sells hygiene products, such as incontinence diapers, feminine hygiene products, and professional hygiene products. The products and services discussed in this interview are incontinence diapers with a service to identify the urination patterns of elderly people in caring homes. The second shortly discussed service is a service for public restrooms. This company provides soap and paper for the public restrooms at for example airports. The service attached to it is a smartphone app which monitors the levels of soap or paper in the dispensers of the restroom. This service is not considered for the analysis of this research, because this is rather a software based service design without people playing a role in delivering the service.

The analysis service for incontinence diapers works as follows. Patients in caring homes have to wear special types of incontinence diapers with built-in sensors inside. These sensors can measure the level of incontinency, which determines the type of diaper they need. This information can also lead to advice to wake someone up at the right time, resulting in that they do not need any incontinence diaper at all. People within the hygiene company were hired to perform analysis and give advice to caring homes. The service consists of a testing phase, an analysis phase, after which the outcomes and recommendations follow. All these steps are modules. See figure 7 for this process.

(32)

32

The first step is the decomposition of the PSS system. The interviewee states that “the care institutions

pay a yearly fee to use the service. The only condition is that they keep buying our products”. Also, the

service is clearly attached to the product which this company provides, because without the incontinence diapers with the built-in sensors, the service could not exist.

Regarding the orientation of the decomposition logics, this analysis service has a combined orientation, because the process steps of this service follow each other sequentially, but could be added or

subtracted if that would be necessary. The interviewee states: “these are steps which have to be taken

sequentially, you start with testing, analysis, outcome, and end with a recommendation. (…) I think it is very difficult, if not impossible to execute these steps in a different order”.

Regarding the levels of the decomposition, the interviewee states: “you can start with a test and you

base your recommendations on this test. (…) It depends per patient what type of product they need, how much it should be able to absorb, how drastic the measures should be that have to be taken for the individual patient. (…) At this moment this is all included in the analysis and recommendations phase, but in my opinion, you could split this up into multiple portions”. As such, this service consists of single level

modules, but could be divided into multiple levels in the future. For this study, however, we consider them as single level modules.

Finally, the identified modularity types are bus modularity and cut-to-fit modularity. The argument for bus modularity is the interviewees’ statement that modules can be added or subtracted, depending on the situation. The interviewee states for instance that “you do not need to perform all the process steps.

(…) Sometimes you only need to do the testing, and decide after that no analysis is needed” (i.e., the

patient is not in need of an incontinence diaper). Cut-to-fit modularity is identified, because each process step can be adapted in duration during the service delivery to fit the customers’ needs. Some patients need more extensive testing than others do.

Interviewee 6 mentions benefits for the customer, such as accessibility of the service for multiple parties due to the division in smaller, cheaper portions. He states: “when you divide the service into multiple

(33)

33

name), these are all healthcare institutions. The government plays an important role for them, and therefore money is an important factor. If the entire service is too expensive for them, you can also offer portions of the service, which is cheaper”. By this he means that technically the company could perform

the analysis phase without the recommendations phase, and instead just send the test results for the healthcare institution to make their own conclusions. Also a wider variety to choose from is an important benefit for the customer according to this interviewee. “if you offer more services (i.e. the service is split up into multiple phases), you can cater to a wider variety of customers, and customers

have more choice options. This is a win-win situation”.

4.7 Interview 7 (i7) – Food industry

The seventh and last interview was conducted with a sales manager from a major food processing company. The previous function of this interviewee was supply chain manager. The interviewees’ responsibilities regard selling cheese products to customers and the corresponding services, which are mainly logistics activities. These logistics services all start with the sale of cheese products, to which several modules of logistics services can be attached, such as basic transport, return flow of pallets, choice in type of transport, choice in type of pallets, and so on. These modules are illustrated in figure 8.

Figure 8 – Logistics services

(34)

34

services. (…) Without the product there is no service”. This is a clear example of a decomposition of the

product-service system as an entity. The customer can choose to buy the product and arrange logistics services themselves, but if they choose any of the logistics activities that this company has to offer, the service will be attached to the product.

Regarding the orientation of the decomposition logics the services are all outcome oriented. The interviewee mentions that it could be a combined orientation, because the customer must buy the product first. According to him this is partly a process, because the sale step comes first. However, this fact is already taken in consideration by taking the decomposition on service or product-service level in account. Therefore the modules that this company offers are outcome modules.

Regarding the levels of decomposition, the interviewee states: “after the cheese has been sold, the

customer may choose from different types of logistics services, such as the type of pallet they want it to be transported with, or the type of transport”. These modules are not divided into smaller portions, and

therefore they are decomposed on a single level.

Lastly, the identified modularity types are sectional modularity and mix modularity. Sectional modularity is identified, because the customer can choose from an unrestricted combination of modules of

transport. These modules can be chosen and combined into any combination. Also, mix modularity is identified, because the service can consist of two or more predefined modules which together create an overall service experience.

Interviewee seven mentions the following benefits for both the company and the customer: “modularity

simply adds transparency for your employees. The employees (i.e. in sales) know exactly which services they are able to sell to the customer”. The customer also benefits from this by knowing exactly what the

company can offer and what the standardized prices are. A disadvantage is that as a company “you don’t

really focus on the customers’ problem anymore. If you only stick to the predefined modules and you’re not able to listen to the customer anymore. You may skip important opportunities to make profit if you ignore customers’ wishes to have a service which is not predefined into modules”. According to the

(35)

35 The company should not confront their customers with their modules, but instead offer a service which appears customized, even though the service may entirely exist of standardized modules.

4.8 Structured benefits from literature

Finally, the interviewees were asked to give their extended view about some of the benefits that modularity has to offer to their product-service systems. The benefits presented to the interviewees were: higher efficiency in offering a product-service system, flexibility, cost savings, customization, economies of scale, and better position to compete by standardizing portions of the service. The descriptive statistics of the ranked benefits of modularity are shown in table 7. It should be said that these statements were presented as benefits of modularity, which they are, so there is a possibility that the rankings were slightly biased towards agreement. However, by comparing the mean scores of the benefits in combination with the variances, it is possible to get a good impression of which benefits were higher agreed upon than others.

N Mean Minimum Maximum Variance

Efficiency 7 4.3 4 5 .2 Flexibility 7 3.7 2 5 1.6 Costsaving 7 4.0 2 5 1.3 Customization 7 3.4 1 5 2.6 Economiesofscale 7 4.0 2 5 1.0 Bettercompetition 7 3.7 2 5 1.2 SuitableforPSS 7 4.7 4 5 .2

Table 7 – Descriptive statistics benefits modularity

4.8.1 Efficiency

Regarding the efficiency benefit the interviewees are in agreement that efficiency is one of the big benefits of modularity. This benefit ranks on a Likert-scale from 1 to 5 an average of 4.3 (n=7) with a variance of 0.2. The interviewees mention that “efficiency depends on the size of the module”. If the modules are too narrowly defined, the company needs too many modules, which defeats the purpose of modularity. Also, one interviewee mentions: “you create efficiency by doing things more often, better,

(36)

36

4.8.2 Flexibility

Regarding the flexibility benefit, the interviewees are slightly less in agreement that this is a benefit of modularity. This benefit ranks on a Likert-scale from 1 to 5 an average of 3.7 (n=7) with a variance of 1.6. One interviewee mentions that “modularity brings the opportunity to tune to customer needs, so in that

sense you are extremely flexible. One other interviewee says that “modularity can create a certain fashion of flexibility which is not desirable”. By this he means that flexibility gives the customer the

choice to downgrade the service in any way he wants, which is not always a desirable option, especially when the company offers a service which is meant to unburden the customer. The customer still expects to be unburdened, while in reality, he is not. An important aspect to flexibility is that, again, the size of the module matters. One interviewee mentions the trade-off between the former mentioned efficiency benefit and flexibility benefit. Flexibility can only be achieved if the modules are not too broadly defined. Another footnote that one interviewee mentions is that “flexibility can increase hugely by means of

modularity, because you’re able to react quicker and more adequate (…) it can also decrease enormously when a company sticks to the modules too rigidly and does not offer any additional choice”.

4.8.3 Cost savings

The interviewees were fairly positive about the benefit of cost savings. This benefit ranks on a Likert-scale from 1 to 5 an average of 4.0 (n=7) with a variance of 1.3. One of the interviewees states that he thinks cost savings are possible due to the opportunity of employing lower-skilled or lower-educated employees. “All-rounders (employees) are hard to find, and are expensive. If you divide the

product-service system into modules, you can tune the skills and abilities of your employees to these modules”.

This is agreed upon by another interviewee who states that modularity can bring huge cost savings because you “remove the human factor by standardizing things. For your best sales guy it will always be

a restriction, but modularity pulls up all those people below the median, because they have a system that they have to comply to. This will always bring cost savings”. Another benefit mentioned by an

(37)

37

4.8.4 Customization

Not all interviewees agreed that customization is necessarily a benefit that stems from modularity. This benefit ranks on a Likert-scale from 1 to 5 an average of 3.4 (n=7) with a variance of 2.6. Some of the interviewees state that customization is actually the opposite modularity, and as such is not a benefit of modularity. “Customization and modularity are each other’s opposites. (…) I think that firms need to

have a ratio 90/10 of modularity and customization, to have a perfect combination”. By this he means

that if a company offers 90% of their product-service offerings in modules, and 10% customization, this results in an optimal catering of the customer’s needs. This interviewee also mentions different types of customers that a company might have. He states “in the business to business market, customization

implies tailor-made services without the offering of modules. In the business to consumer market you can offer tailor-made services by offering modules to the customer. Every car exists of modules, and still people think they have their own custom made vehicle. The business to business market does not work like that”.

4.8.5 Economies of scale

Regarding the economies of scale benefit, the interviewees do think that this is a benefit of modularity. This benefit ranks on a Likert-scale from 1 to 5 an average of 4.0 (n=7) with a variance of 1.0. Two interviewees state that you can achieve economies of scale because of more efficient purchasing of services that come along with the products. An interviewee states that “by making the entire

product-service combination modular, the predictability (through standardization) of delivering these products and services increases. You can get economies of scale from that, because (…) you can get better prices if you are able to predict what you are going to deliver in the future”. Another interviewee states: “we have a 3R department, repair, refit, and refurbish. These are modernization services. We don’t want one specialist per discipline who’s idle half of the time”. By this he means that through modularity (as in

standardization), less specialists are needed to perform the same job.

4.8.6 Better position to compete by standardizing parts of the service

(38)

38 of the interviewees disagrees with this statement, because “by standardizing your services, you make it

very easy for your competitors to imitate your services”, which is not very beneficial for your ability to

compete. Another one does agree with the statement, because “you’re able to react quicker to customer

demands, especially when the customer asks for a price enquiry”.

4.8.7 Modularity is suitable for product-service systems

(39)

39

5 Discussion

Below each of the sub-research questions will be revisited. First the basic choice in decomposition of the PSS is reflected upon, followed by the decomposition logics (including decomposition orientation and decomposition level), the recognized modularity types, and the perceived benefits and disadvantages of modularity.

5.1 Basic choice in decomposition

The first sub-research question was: What is the basic choice in decomposing PSSs? Here a basic design choice was made whether the service should be decomposed separately, or together with the product. As shown in table 8, there are four services which are decomposed separately, and four services which are decomposed as product-service system (i.e. together with the product). Upon first sight, the results seem scattered, but it becomes clear that all the services that have something to do with maintenance (such as modernization, and repair and maintenance) are decomposed separately from the product. An explanation for this is that the service that these manufacturing firms provide can be applied to

products from their competitors as well. As such, customers with products from competitors also offer their products for maintenance at the interviewed product-service providers.

Table 8 – PSS decomposition basic choice

Company (service) Services seperately decomposed PSS decomposed into smaller PSS parts

Interview 1 (DfX)

Interview 2 (modernization)

Interview 2 (training and documentation)

Interview 3 (repair and maintenance)

Interview 6 (analysis service)

(40)

40

5.2 Decomposition orientation (decomposition logics)

Research question 2a was: What is the decomposition orientation of PSS-providers? Here the

interviewees could indicate whether their service modules were process oriented, outcome oriented or a combination of these two orientations. As shown in table 9, the results are scattered. In one case the modules followed a process oriented decomposition, in four cases the modules followed an outcome oriented decomposition, and finally in three cases the modules followed a combined oriented

decomposition. One thing that is notable from these results is, because a combined orientation consists of both outcome and process modules, an outcome orientation seems dominant in the decomposition of product-service systems.

Company (service) Process oriented Outcome oriented Combined orientation

Interview 1 (DfX)

Interview 7 (logistics services)

Table 9 – Decomposition orientation

5.3 Decomposition levels (decomposition logics)

Research question 2b was: Which decomposition levels do PSS-providers want to decompose their

product-service combinations on? Here the interviewees could indicate whether the recognized modules

(41)

41

Company (service) Single level decomposition Multilevel decomposition

Interview 1 (DfX)

Interview 7 (logistics services)

Table 10 – Decomposition levels

5.4 Decomposition logics

The second sub-research question was: Which decomposition logics can be recognized in PSS offerings? The answering of this sub-research question consisted of two parts as described above: the

decomposition orientation and the decomposition levels. When sub-research question 2a and 2b are combined, the decomposition logics of product-service providers can be identified. These are shown in table 11. Again, the results seem scattered. This study failed to identify any services with a process orientation which are decomposed on multiple levels.

Outcome-oriented Process-oriented Combined orientation

Single level

Modernization (i2)

Repair and maintenance (i5) Logistics services (i7)

Repair and maintenance (i4) Analysis service (i6)

Multilevel Repair and maintenance (i3) DfX (i1)

Training and documentation (i2)

Table 11 – Decomposition logics

5.5 Modularity types

The third sub-research question was: Which modularity types can be recognized in product-service

systems? The eight services are assessed on the type of modularity that follows from the decomposition.

Modular decomposition logics and the perceived benefits of applying modularity in product-service systems