A practical knowledge management approach

(1)

Effectiveness through managing knowledge

A practical knowledge management approach

Sebastiaan van Agteren MScBA, 1166891

(2)

Effectiveness through managing knowledge: a practical knowledge management approach

August 29^th 2005

Author: Sebastiaan van Agteren MScBA, 1166891 Supervisor: Prof. Dr. Ir. A.C.J. de Leeuw

Co-reader: Prof. Dr. Ir. J.C. Wortmann Master thesis

Department of Business Development Faculty of Management and Organization University of Groningen

The Netherlands

The author is responsible for the contents of this master thesis. The copyright of this master thesis lies at the author.

(3)

Preface

Going forward and backward at the same time Prof. Dr. De Leeuw

This paper is the master thesis for the master Business Development of the faculty Management and Organization at the University of Groningen, the Netherlands. The origins of this paper are twofold. The primary origin is the knowledge related problem ASML had with the repair process of their malfunctioning machines. This problem has been investigated is a 5-month research period. The second origin of this paper lies in my master thesis in Strategic Management at the Erasmus University, Rotterdam, the Netherlands. In this research I have constructed a mainly theoretical knowledge strategy framework. For solving the organisational problem at ASML, this framework has been applied.

I would like to thank everybody who contributed to this research. First of all I especially would like to thank my supervisor Prof. Dr. De Leeuw and co-reader Prof. Dr. Wortmann for their support during the writing of this paper. They helped me to sharpen my research question and they provided many valuable comments with much constructive criticism.

Secondly, I am very grateful to ASML for giving me access to their organization. In particular I would like to thank Geert Jan van Herwaarden for introducing me at ASML.

Furthermore I would like to thank my company coach Rogier Kuijpers, my operational supervisor Gerald Fredriksz, my very kind office mates, and everyone at ASML who generously helped me with my research. Finally, I would like to thank everyone at ASML in Dresden for their willingness to contribute to this research.

Last but not least I would like to thank my parents and brother, my girlfriend, my family, and friends who have believed in me and who have supported me throughout my study period in general and with this master thesis in particular.

Haarlem, August 29^th 2005 Sebastiaan van Agteren

(4)

Executive summary

The objectives of this research are twofold. The main and first part of this research consists of solving an organizational problem at ASML. In the second part of this research the experiences of the research have been confronted with the knowledge management literature in order to confirm theoretical insights.

The main and first part if this research consists of the investigation of an organizational problem at ASML in a five-month research period. ASML is a high-tech and knowledge intensive firm. The goal of this research was to improve the efficiency in which breakdowns on the expensive and complex lithographic machines in the field are solved.

These breakdowns, called the unscheduled down process, needed to be reduced in order to meet increased customer requirements.

The organisational problem has been investigated with help of the approach by De Leeuw (1996) which consists of three distinct phases: diagnosis, design, and change. Due to research restrictions only the phase’s diagnosis and design have been elaborated.

Recommendations for the unscheduled down process have been given from a knowledge management perspective. This has been done for two reasons. First, the organizational problem at ASML is mainly knowledge management related. Second, the interest and the expertise of the author focus on knowledge management.

In order to solve the organisational problem, the following main research question has been composed:

Main research question: What recommendations can be given to improve the unscheduled down process from a knowledge management perspective?

In the first step in answering the main research question the unscheduled down process has been diagnosed:

Sub-question 1a: How can the unscheduled down process be diagnosed from a knowledge management perspective?

The diagnosis showed that the unscheduled down process is truly problematic and that it could best be investigated by a combination of an organisational and a knowledge management perspective. When evaluating the problem it came forward that all actors of the unscheduled down process are worried about the (lack of) use of knowledge (elements) in the unscheduled down process. Furthermore, the organisational problem is dispersed throughout the organisation, and there is a big divide between the formal and informal organisation. With these observations in mind, the unscheduled down process has been described by discussing the process, local learning, and organisational learning (see figure 1).

Figure 1 Link between process, local learning, and organisational learning

Process (1) Local

learning (2)

Organisational learning (3)

(7)

The description of the process (1) showed that there are important differences between the actual process and the way it is managed. This comes forward in the role of the customer, in the predictability, and in the evaluation possibilities of the process. This implies that the process is not well managed. The description of the local learning (2) discussed the knowledge element experience, which consists of learning by doing and training. This showed that learning by doing is particularly problematic for the engineers who operate the machines in the fabrication site (the so-called FSE’s). Furthermore, the training programs do not present sufficient skills for solving unscheduled downs. The organisational learning section (3) discussed the knowledge elements cases, tools, and methods. This showed that their quality is so low that they are not used.

In the second sub-question recommendations have been constructed of how the unscheduled down process can be improved with help of a knowledge management perspective:

Sub-question 1b: What recommendations for improvement can be given in order to improve the unscheduled down process from a knowledge management perspective?

From the diagnosis the sub-question has been translated into a concrete management problem statement: How can the use of knowledge elements in the unscheduled down process be optimized in such a way that the organizational problem in the problem area is solved? The main problem statement has been answered by first identifying directions for recommendations by discussing helpful literature and confronting this with the diagnosis. From this three knowledge management directions have been identified for giving recommendations: training, management, and quality of knowledge elements.

Additional training ensures that the engineers have the most necessary skills for solving unscheduled downs. Proper management of the use of knowledge elements is necessary for the engineers to actually use their newly acquired skills. The development of knowledge elements is important in order to ensure that the quality of knowledge elements is sufficient to be used. Next to knowledge management recommendations also organisational recommendations have been presented which advise to improve the unscheduled down process by improving interfirm management in the fabrication sites.

In the second part of this research, the experiences of the research have been confronted with the knowledge management literature in order to confirm theoretical insights.

Sub-question 1c: What insights for the knowledge management literature can be derived from the research at ASML?

The insights for the knowledge management literature have been acquired by discussing the applied knowledge management concepts knowledge elements, local versus organisational learning, and knowledge strategies. First, the concept of knowledge elements proved to be highly useful for unravelling a complex set of knowledge streams.

Second, the distinction between local and organisational learning was useful for emphasizing the crucial difference between tacit versus explicit knowledge. Finally, the knowledge strategy framework’s usefulness is fairly high since it resulted in practical recommendations, although there are shortcomings. The addition of designing of knowledge strategies was useful and, although there are important limitations, it seems an interesting topic for further research.

(8)

Chapter 1: Introduction

The objectives of this research can be subdivided in two parts. The main and first part of this research consists of solving an organizational problem at ASML. In the second part of this research the experiences of the research will be confronted with the applied knowledge management literature for additional insights.

The main and first part of this research consists of the investigation of an organizational problem at ASML in a five-month period. ASML is a world leader in the manufacturing of advanced technology lithographic systems for the semiconductor industry. The company offers an integrated portfolio for manufacturing complex integrated circuits (also called ICs or chips). The ASML machines are constructed in a modular way: they consist of multiple sub-systems that are relatively independent from each other. These machines are very expensive (about € 20 million a piece) and machines failures (the so- called unscheduled downs) cost the customer between € 10.000 to € 50.000 per hour. The semiconductor industry market is becoming increasingly cost-oriented, which makes the customer and ASML more and more cost-oriented. This has resulted in the demand by the customer for a more efficient unscheduled down process and the decision of ASML to investigate the possibilities for this.

For investigating the unscheduled down process a knowledge management perspective has been chosen. The reasons for choosing a knowledge management perspective are twofold. The first reason is that the organizational problem at ASML is knowledge management related. The second reason is that the interest and the expertise of the author of this investigation lie on the topic of knowledge management.

In the second part of this research, the lessons from the research at ASML will be used to look at the knowledge management literature. By seeing the five-month investigation as a case study the insights can be used for improving the knowledge management literature.

The contributions of this research are thus twofold. From a managerial point of view the most important insights will be given on how a complex knowledge related problem can be solved at ASML. However, other organisations can also benefit of this research by getting a good feel of the complexity of this organisational problem. Secondly, theoretical contributions will be presented by reflecting the insights of the investigation at ASML with the applied knowledge management literature.

(9)

Chapter 2: Research approach

Introduction

In the research approach of this research first a global discussion of the research process will be presented. In the research goal the main research goal and sub-questions are described. Next, the data sources & measure and observation methods will be discussed.

Finally of an overview of the investigation is presented.

Research process

For solving managerial problems the approach by De Leeuw (1996) is very appropriate.

In this book he classifies business research in three activities: diagnosis, design, and change (see figure 2.1 below). Due to feasibility reasons it has not been possible to perform all three activities: in this research only the diagnosis and design phase are treated. Thus, implementation of the design falls out of the scope of this research.

Although the research process is pictured as a sequential process, it is possible that in reality this is an iterative process. The research approach can be characterised as being multidisciplinary and highly suitable for the subjective reality of management.

Research goal

The goal of this investigation is to give recommendations to ASML on how the time of breakdowns on their machines (so-called unscheduled downs) can be reduced in order to reduce Labour Hours per Machine (LHM) and to increase machine availability. This will be done by diagnosing, and giving recommendations for improvements of the unscheduled down process.

(10)

Since it is believed that the organizational problem is highly knowledge related, a knowledge management perspective is chosen. The main research question of the case study is thus:

Main research question: What recommendations can be given to improve the unscheduled down process from a knowledge management perspective?

The first sub-question of this research attempts to diagnose the current process of solving unscheduled downs and its organizational problem from a knowledge management perspective:

Sub-question 1a: How can the unscheduled down process be diagnosed from a knowledge management perspective?

The second sub-question shall construct recommendations of how the unscheduled down process can be improved with help of a knowledge management perspective:

Sub-question 1b: What recommendations for improvement can be given in order to improve the unscheduled down process from a knowledge management perspective?

Next to managerial implications, the five-month investigation at ASML can also be seen as a case study. Seen in this light it may also have important insights for the knowledge management literature. This will be investigated in the following final sub-question of this research:

Sub-question 1c: What insights for the knowledge management literature can be derived from the research at ASML?

Data sources & measure and observation methods

The data sources of this research are both quantitative and qualitative, but the emphasis lies on qualitative data. The qualitative data can be subdivided into several sources. In the beginning of the research at ASML many informal discussions and meetings have been held in order to get a good overview of the background and problem area of the research.

The second important source of qualitative data has been a ten-day visit to the regional office and the customer fabrication site in Dresden, which consists of observations and interviews. Finally, during the five-month internship much informal communication with relevant actors has been an important data source. Because at ASML there is a big distinction between the actual and formal reality, interviews with key players is seen as the most important source of information of this research. Relying too much on formal documents may give a wrong impression of reality. The immediate cause of this investigation is a sense of urgency by the management, which is backed by quantitative data from the ASML Customer Support Equipment Performance Management department.

Overview of the investigation

After the description of the introduction (chapter 1) and the research approach (chapter 2) in chapter 3 the organisational problem will be diagnosed. Based on the diagnosis in chapter 4 design and recommendations will be constructed for solving the organisational problem. In chapter 5 will be investigated to which degree the experiences at ASML can be used for insights in the knowledge management literature.

(11)

Finally, conclusions, and recommendations will be given in chapter 6. This research process is depicted in figure 2.2.

Figure 2.2 Research process

Chapter 6:

Conclusions, and recommendations Chapter 5:

Insights knowledge management literature Chapter 4:

Design and recommendations Chapter 3:

Diagnosis Chapter 2:

Research approach Chapter 1:

Introduction

(12)

Chapter 3: Diagnosis

Introduction

In the diagnosis the following sub-question of this research will be answered: how can the unscheduled down process be diagnosed from a knowledge management perspective?

In order to answer this sub-question five different parts of the diagnosis will be discussed (see figure 3.1). First a diagnostic problem statement is constructed where the initial cause of the investigation is described. This is the starting point of the diagnosis.

Following in the multiform analysis two promising perspectives are selected for describing and analyzing the organisational problem. Next is evaluated in which way and to which degree the process of solving unscheduled downs is problematic. In the description the process of solving unscheduled downs at ASML is constructed. In the concluding section the main problem statement for the design phase is composed out of the previous four diagnosis sections.

Figure 3.1 Framework diagnosis (De Leeuw, 1996) 3.1 DIAGNOSTIC PROBLEM STATEMENT

Here the initial cause of the investigation will be described. This is important for optimally understanding the organisational problem. The initial cause of the investigation will be investigated by introducting ASML and its machines, problems with breakdowns on these machines, and initial cause for starting this research.

ASML is a world leader in the manufacturing of advanced technology lithographic systems for the semiconductor industry. The company offers an integrated portfolio for manufacturing complex integrated circuits (also called ICs or chips). The ASML machines are constructed in a modular way: they consist of multiple sub-systems that are relatively independent from each other. These machines are very expensive (about € 20 million a piece) and machines failures (the so-called unscheduled downs) cost the customer between € 10.000 to € 50.000 per hour. See for an overview of the breakdown of the total machine time figure 3.2.

Because machine failures are so costly, it is imperative to meet the specifications of the availability of the machines. Furthermore, long unscheduled downs lead to a higher Labour Hours per Machine (LHM) than necessary. LHM are directly related to the ASML performance towards the customer, since the LHM are recorded in performance contracts. The optimisation of the availability of the advanced technology systems is done by the Customer Support department at the ASML headquarters.

Diagnostic problem statement Diagnosis

Evaluation

Description Multiform

analysis

Main problem statement

(13)

The availability of machines depends on two factors: the Mean Time Between Interrupts (MTBI) and on the Mean Time To Repair (MTTR).

Figure 3.2 Subdivision of total machine time

A tendency can be observed that in the semiconductor industry the market is changing from an innovative market towards a more cost-oriented market. This tendency has made the ASML customers more and more cost conscious. The result is that the customers no longer accept a high unscheduled down time and push ASML to reduce this time to a minimum. Within ASML the unscheduled down time and MTTR have become very important Key Performance Indicators (KPI’s). The manager of the sector Customer Support – Product Generation Engineering, has decided that it should be investigated how the current unscheduled down process can be improved.

How the unscheduled down process must be improved will be further investigated in the remainder of this chapter.

(14)

3.2 MULTIFORM ANALYSIS Introduction

From the diagnostic problem statement comes forward that the unscheduled down process is very expensive and that there is an increasing pressure from the customer and ASML to improve this process. For ASML management this has become an organisational problem which needs to be resolved. In order to get a good view on the organisational problem and in order to answer the main research question first a multiform analysis will be conducted. According to De Leeuw (1996) a multiform analysis is conducted in order to identify multiple promising theoretical perspectives for describing and improving the organisational problem. The advantages of multiple theoretical perspectives are that by looking from different perspectives the problem can be better diagnosed and solved, and that the chance on sub-optimal solutions is minimized.

In order to identify promising theoretical perspectives for improving the unscheduled down process, it is important to identify what initiatives and perspectives have already been used by ASML for improving this process. At the Customer Support (CS) department several initiatives have been undertaken in order to reduce the unscheduled down process. For example, in the beginning of 2003 a project has been started to improve the availability and access to the various on line databases through the Intranet.

The project (called SAP CRM phase 1) uses an integrated Intranet portal in which many different databases are grouped. Through making the relevant information uniform available and accessible it was believed that the unscheduled down process could be reduced. Phase 1 of the SAP CRM portal will be on-line and available for the FSE’s in the field in the beginning of 2006. Another initiative which attempts to have an impact on the unscheduled down process is set up by CS CSE Product Generation Engineering department. From this department originates the Diagnostics Roadmap which is the vision of how diagnostics on the machines should be organized in the future. The roadmap can be characterized as attempting to standardize the unscheduled down process by automating the process as much as possible.

From the description of the organisational problem and the discussion of the ASML initiatives the following two theoretical perspectives have been identified.

The first perspective is a generic organisational perspective, which focuses on the effectiveness and efficiency of the unscheduled down solving process overall. The reason why is chosen for this perspective is that the problems with the unscheduled down process seem to be very broad. The solution of this problem might have an impact on the topics such as strategy, structure, and processes.

The second perspective is a knowledge management perspective and it particularly focuses on the knowledge demand and learning opportunities of the unscheduled down solving process. Previous initiatives at ASML to reduce unscheduled down process have been ICT-focused. It is assumed here that the unscheduled down process problems is primarily knowledge related. Providing the right information at the right time is only a part of the solution. To which degree the information is used by the FSE’s and the process in which the unscheduled down process is solved is critical. This problem can be tackled with help of knowledge management.

(15)

3.2.1 ORGANISATIONAL PERSPECTIVE

From an organisational perspective the organisational processes will be assessed in order to determine if the solving of unscheduled downs is efficiently and effectively organised.

Five theories have been selected for describing the organisational perspective. These are subsequently the criteria for evaluating organisational performance, the distinction between structured and unstructured problems, organisational structure, coordination of interdependence, and the organisational boundary.

Criteria for evaluating organisational performance

Evaluation criteria for organisational performance is necessary for assessing to which degree the organisational problem at ASML is problematic. In the organisational literature Haselhoff (1977) identified three criteria on which organisational performance can be evaluated. These criteria are effectiveness, efficiency, and social meaning.

Effectiveness is important in order to assess if the right actions are taken. Efficiency is important in order to evaluate if the taken actions are performed at a minimal effort.

Social meaning, finally, is important because the people who perform actions need to have a certain degree of work happiness in order to function properly in the long run.

Social meaning is particularly important when the employees need to share their own knowledge in the organisation.

Structured versus unstructured problems

The distinction between structured and unstructured problems is useful for better understanding the nature of the organisational problem. In his classic work

‘Administrative Behaviour’ Simon (1977) touches upon many important topics. Two of these will be elaborated here because it is believed that these are important for this research. The first one is concept of bounded rationality. This means that there is a difference between rational decision making and managerial practice. Another topic is the distinction between structured and unstructured problems. Structured problems are problems in which most elements are known and which can be solved routinely.

Unstructured problems on the other hand have a high degree of uncertainty and must be solved by innovative decisions.

Organizational structure

Organizational structure is a very well-known element of the organizational design. The topic of organizational structure will be regarded carefully in this research, since a knowledge domain mostly is only a part of an organization and thus there are limits to which a knowledge domain can be redesigned. Thompson (1967) defines organizational structure as the internal differentiation and patterning of relationships of the major components of a complex organization (such as departments). Mintzberg (1979) has about the same definition of organizational structure as Thompson. Mintzberg (1979) defines organizational structure as the sum total of ways in which it divides its labour into district tasks and then achieves coordination among them. He has identified five fundamental ways (mutual adjustment, direct supervision, standardization of work processes, standardization of work outputs, and standardization of worker skills) in which organizations coordinate their work, of which he argues that they should be considered as the most basic elements of structure.

(16)

1. Mutual adjustment achieves coordination of work by the simple process of informal communication.

2. Direct supervision achieves coordination by having one person take responsibility for the work of others, issuing instructions to them and monitoring their actions

3. Standardization of work process occurs when the contents of the work are specified or programmed

4. Standardisation of output comes forward when the results of the work – for example, the dimensions of the output or the performance – are specified.

5. Standardisation of skills (and knowledge) takes place when the kind of training required to perform the work is specified.

Coordination of interdependence

Understanding the interdependences in a particular process is important, because the degree of interdependence has important consequences for management and organisational design. Thompson (1967) states that the main problem of organisations is dealing with uncertainty. Uncertainty can be reduced by subdividing tasks in parts which have a low uncertainty. The primary sources of uncertainty are the used technology and the environment. Organisations and particularly its core technology should be as less as possible dependent of the environment. According to Thompson (1967) technology should be defended by organising input and output activities in three ways. First, loosely coupled activities (pooled activities) should be organised by standardisation where possible. Sequential activities should be absorbed by the planning of activities. Activities which are reciprocal interdependent cannot be standardized or planned, but must be organised by mutual adjustment.

Organisational boundaries

A discussion of the boundary literature is useful for marking out the organisational problem and better understanding its dependencies. In the article of Foss (2001) the boundary school literature and its most important approaches are discussed. In this section the approaches transaction cost economics, resource dependency approach and the capabilities approach shall shortly be discussed. The most important aspects of these approaches are summarised in table 3.1.

1. Transaction cost economics

In this classic boundary school approach the focus lies on the transaction between organisations and the focus lies on the efficient organisation of transactions and assets.

Transactions can mainly be done by markets or hierarchies. In the choice between the two asset specificity is key. Efficient boundaries are understood in terms of incentives and transaction costs which are related to markets and hierarchies.

2. Resource dependency approach

In this approach the dependence relation between an organisation and its boundary and the control over and dependence on resources is central.

In this approach it is assumed that organisations attempt to reduce their dependency on resources as much as possible. Furthermore, there may be an important distinction between formal ownership and the actual control in practice.

(17)

3. Capabilities approach

In this approach the focus lies on capabilities in the sense that capabilities may lead to competitive advantage and that control over capabilities may lead to knowledge accumulation. This depends on the degree of complementarity and similarity of capabilities.

Transaction cost

economics

Resource dependency approach

Capabilities approach Main trust The efficient

organisation of transactions and assets

The control over and dependence on resources

Competitive advantage;

knowledge accumulation Unit of analysis The transaction The dependence relation Capabilities

Understanding of boundaries

Asset specificity is key;

efficient boundaries understood in terms of incentives and transaction costs

Attempts to reduce dependence; contrast between formal ownership and control

Depends on the degree of complementarity

important and similarity of capabilities

Table 3.1 Taxonomy of approaches constituting the boundary school (Foss, 2001) 3.2.2 KNOWLEDGE MANAGEMENT PERSPECTIVE

From the description of the subject and the goal of the investigation it became clear that knowledge plays an important role in the unscheduled down process. Much information is available for solving unscheduled downs but it is felt that but this is not optimally applied. Therefore a knowledge management perspective is chosen in order to investigate the opportunities for improving the unscheduled down process. The knowledge management perspective will be elaborated by subsequently discussing knowledge, knowledge elements, local versus organisational learning, knowledge management, and knowledge strategies.

Data, information, knowledge

One widespread classification of the concept of knowledge is the distinction between data, information, and knowledge. According to Davenport (1998) data is a set of discrete, objective facts about events. There is no inherent meaning in data and data says nothing about its own importance or irrelevance. Data is important to organisations because it is essential raw material for the creation of information. Davenport (1998) defines information as a message, usually in the form of a document or an audible or visible communication, which has a sender and a receiver. The primary goal of information is to inform. Data becomes information when its creator adds meaning. Data can be transformed into information by adding value in various ways. Contextualizing, categorising, calculating, correcting, and condensing can do this. When defining knowledge Davenport (1998) offers a working definition of knowledge, which expresses the characteristics, that make knowledge valuable and that make it difficult to manage as well. According to Davenport (1998) knowledge is a fluid mix of framed experience, values, contextual information, and expert insight that provides a framework for evaluating and incorporating new experiences and information. It originates and is applied in the mind of the individuals who know. In organisations, it often becomes embedded not only in documents or repositories but also in organisational routines, processes, practice, and norms. Knowledge derives from information as information derives from data.

(18)

The transformation from information to knowledge happens through comparison, consequences, connections, and conversation. Knowledge-creating activities take place within and between humans. According to Nonaka (1994) the distinction between information and knowledge can be seen as follows. Information is a flow of messages, while knowledge is created and organised by the very flow of information, anchored on the commitment and beliefs of its holder.

Tacit and explicit knowledge

A second classification of the concept of knowledge is the distinction between tacit and explicit knowledge. Tacit knowledge is deeply rooted in individual’s actions and experience (Koskinen et al., 2003). Also, the ideals, norms and values or emotions that peoples hold form input for tacit knowledge. The fact that tacit knowledge is hard to define and quite elusive is enhanced by the fact that subjective insights and hunches are also considered as parts of tacit knowledge (Koskinen et al., 2003). Tacit knowledge is personalised, it resides in the heads of people and cannot easily be codified. Explicit or articulated knowledge can take on a concrete form, such as a written text, words, databases, and documents (Davenport & Prusak, 1998). Polanyi (1996) makes a distinction between explicit and implicit knowledge. According to Polanyi implicit is the shared sense of ‘what we do with what we know’. Implicit knowledge has fairly the same meaning as tacit knowledge. According to Osterloh & Frey (2000) the distinction between the two types of knowledge is important because of the transferability and appropriability of explicit knowledge, as opposed to tacit knowledge (Grant, 1996). Tacit knowledge is acquired by and stored within individuals and cannot be transferred or traded as a separate entity. Explicit knowledge has the character of a public good. Two important consequences follow. First, tacit knowledge is a crucial source of sustainable competitive advantage because it is difficult for competitors to imitate it (e.g. Teece, 1997). Second, the contribution of a particular employee’s tacit knowledge to a team output cannot be measured and paid accordingly. This has important motivational ramifications (Osterloh & Frey, 2000).

In his article ‘Misperceptions on managing knowledge’ Oldenkamp (2002) is critical on the distinction between tacit and explicit knowledge and calls it arbitrary. In his article he maintains the definition of knowledge by Davenport (1998). When explicit knowledge is defined as knowledge that is able to take a concrete form it is no longer knowledge but information. Despite of the importance of this argument, the distinction between explicit and tacit knowledge is considered important. The first reason for this is that the distinction indicates that not all knowledge is the same. Some knowledge can be more easily shared than other knowledge. The second reason is that the distinction is widely used in the literature and that it thereby connects to a wide research stream.

Knowledge elements

Werr & Stjernberg (2003) have identified three interacting knowledge elements, which are important in knowledge intensive firms: methods and tools, cases, and experience (see table 3.1). Knowledge elements are the means by which knowledge can be exchanged in organisations. Experience is an example of tacit knowledge, and cases, tools, and methods are examples of explicit knowledge.

(19)

Table 3.2 Three interacting knowledge elements

Methods and tools are articulated, organisational and abstract knowledge that is widely applicable, not very specific, and easily transferable. Cases are documents, which are articulated, organisational, but specific knowledge and which are limited general applicable, useful in similar cases, and easily transferable. Experience is tacit, individual and has varying levels of abstraction. Experience guides actions, is hard to transfer, and for transfer face-to-face interaction is needed. Experience can be increased by learning- by-doing and by training. These three knowledge elements interact with each other on two dimensions (see figure 3.3).

Figure 3.3 Interaction knowledge elements

(20)

The first dimension is degree of tacitness. Experience is very tacit whereas methods and tools, and cases are more explicit. The second dimension is the degree of specificity.

Cases require very specific knowledge, whereas methods require very general knowledge: experience lies in this continuum in between.

From the discussion of the various aspects of knowledge and the theory of knowledge elements the following working definition of knowledge is:

Applied knowledge elements by individuals Local versus organisational learning

From the discussion of the knowledge elements it came forward that there is an important difference between tacit (experience) and explicit knowledge (cases, tools, and methods).

An important aspect of tacit knowledge is that it can only be transferred in a social context. The result is that experience (tacit knowledge) is per definition local. An increase of experience by learning-by-doing & training can therefore be seen as local learning.

The use and improvement of the explicit knowledge elements cases, tools, and methods can be typified as organisational learning. The link between a specific process, local learning, and organisational learning can be seen in figure 3.4.

Figure 3.4 Link between process, local learning, and organisational learning

As can be seen from the figure there is interaction between a specific process and local learning, and between local learning and organisational learning. When a specific process is performed the experience of an individual shall increase (1). This leads to local learning. On the other side, increased experience shall result in an improved process (2).

In the link between local learning and organisational learning the link between explicit knowledge elements and tacit experience is key. Explicit knowledge elements from the field must be used in order to enrich experience (3). This is a critical step for two reasons.

First individuals tend to trust their own experience more than explicit knowledge. Second the application of explicit knowledge elements may require a different way of working, which often is a barrier. Finally, explicit knowledge must be renewed with increased experience of individuals (4). This also is a critical step, since this is an additional task to perform. Individuals must be stimulated and possibly rewarded for this.

Knowledge management

From the discussion of the concepts knowledge, knowledge elements, and the difference between local and organisational learning, now the definition of knowledge management will be given.

Sarvary (1999) defines knowledge management as a business process in which organisations create and use their institutional or collective knowledge. He has identified three sub-processes: organisational learning, knowledge production, and knowledge distribution.

3

2

Process 1 Local

learning

Organisational learning 4

(21)

Organisational learning is the process through which the firm acquires information and/

or knowledge. Knowledge production is the process that transforms and integrates raw information into knowledge that in turn is useful to solve business problems. Knowledge distribution is the process that allows members of the organisation to access and use the collective knowledge of the firm.

In this research another definition of knowledge management is used. For effective knowledge management all knowledge elements must be managed. It is assumed that adapting a learning cycle, such as the learning cycle by Kolb, is suitable for this. In this investigation a learning cycle that exists from the phases development, introduction, monitoring, and improvement is used (see figure 3.5). An essential addition to the knowledge management definition is that the knowledge must be managed in such a way, that it is actually used. The reason for this is that when knowledge elements are not used, no learning takes place and the knowledge elements are useless.

The definition of knowledge management in this research is thus:

The development, introduction, monitoring, and improvement of knowledge elements on a specific topic in such a way that they are applied.

Figure 3.5 Complete knowledge management process

Knowledge strategies

As already noted, knowledge is seen as a critical resource in the organization, but knowledge cannot be managed in isolation. In order to efficiently pursue an organizational strategy, all resources within the organization should be in fit in order to effectively and efficiently follow a particular strategy. Hansen, Nohria & Tierney (1999), for example, argue that knowledge management should have close connections with functional departments such as human resource management and IT, and competitive strategy. Zack (1999) emphasizes the importance of the link between the business strategy and the knowledge strategy. Zack stresses the importance to either align strategy to what the organization knows or developing the knowledge and capabilities needed to support a desired strategy. As organisations can pursue multiple organisational strategies, also multiple knowledge strategies can be identified. Van Agteren (2004) has constructed a knowledge strategy framework (see figure 3.6) in which the choice of a knowledge strategy depends on the environmental dynamism and maturity of a knowledge domain.

Management of knowledge elements

Development

Improvement

Monitoring

Introduction Knowledge

elements

(22)

In stable environments, for example, the knowledge strategy should focus much more on standardizing knowledge than in dynamic environments. And young knowledge domains shall focus much more on creating knowledge than mature knowledge domains. The knowledge strategy framework and its elements knowledge domains, dynamism, maturity, dynamic capability types, and the four knowledge strategies will be elaborated here.

Figure 3.6 Knowledge strategy framework Knowledge domains

The unit of analysis of the knowledge strategy framework is the so-called knowledge domain and not the organisation as a whole. It is believed that within organisations multiple knowledge domains can be identified which can have different knowledge strategies. This is the reason why the knowledge domain is chosen as the unit of analysis.

Knowledge domains consists of relevant data, information, articulated knowledge, and a list of key people and groups with tacit knowledge based on long-term work experience on a certain topic) in order to achieve a certain organizational goal (Von Krogh et al., 2001). In short, knowledge domains consist of two elements: the collection of knowledge elements on a particular topic and group of people unified in a so-called Community of Practice on the same topic.

Achieving a certain organisational goal is done in cooperation with experts within the knowledge domain who are unified in a Community of Practice. A Community of Practice consist of a core group (usually 10 to 12 members) and is expanded by the inclusion of other people in the same line of practice and which shall contribute to an identified knowledge domain (Von Krogh et al., 2001).

Moderately dynamic environment High velocity

dynamic environment

Mature knowledge domain Young knowledge domain Expanding strategy

o Gaining o Releasing o Reconfiguring

Applying strategy o Reconfiguring

Appropriating strategy o Gaining o Reconfiguring Innovating strategy o Gaining o Integrating

(23)

The purpose of Communities of Practice is to act as a custodian for the knowledge domain, nurturing the sharing and creation of practices and knowledge that is key to the achievement of both company and personal objectives. In daily life people interact in various relationships and groups. In these relationships and groups, people interact and tune relationships with each other and thus learn. This collective learning results in practices that reflect both the pursuit of our enterprises and the attendant social relations.

These practices are the property of a kind of community created over time by the sustained pursuit of a shared enterprise. This is what Wenger (1998) calls Communities of Practice. In this definition the list of key people and groups with tacit knowledge on a specific topic in the definition of knowledge domains by Von Krogh et al. (2001) can be typified as a Community of Practice.

Dynamism

The first dimension of the knowledge strategy framework is the degree of dynamism of the knowledge domain. According to Zollo & Winter (2002) the definition by Teece et al.

seems to require the presence of ‘rapidly changing environments’, but firms obviously do integrate, build, and reconfigure their competencies even in environments subject to lower rates of change.

In order to understand fully dynamism, the concept of dynamic capabilities will be addressed. In markets, where the competitive landscape is shifting, the dynamic capabilities by which firm managers integrate, build, and reconfigure internal and external competencies to address rapidly changing environments become the source of sustained competitive advantage (Teece et al., 1997). As such, they are the drivers behind the creation, evolution, and recombination of other resources into new sources of competitive advantage (Teece et al., 1997). Dynamic capabilities, as defined by Teece et al. (1997), do not involve production of a good or provision of a marketable service.

Instead, as noted above, dynamic capabilities build, integrate, or reconfigure capabilities.

Thus, dynamic capabilities do not directly affect the output for the firm in which they reside, but indirectly contribute to the output of the firm through an impact on operational capabilities (Helfat & Perteraf, 2003). The type of dynamic capabilities differs depending on the market dynamism.

In particular, dynamic capabilities vary in their reliance on existing knowledge.

Moderately dynamic markets are ones in which change occurs, but along roughly predictable and linear paths. The industry structure is stable and there are clear boundaries and competition. These markets rely heavily on existing knowledge. In moderately dynamic markets, dynamic capabilities resemble the traditional conception of routines and effective dynamics capabilities are detailed and analytical. In contrast, when markets are very dynamic or what is termed high velocity (e.g., Eisenhardt, 1989), change becomes nonlinear and less predictable. High-velocity markets are ones in which market boundaries are blurred, successful business models are unclear, and market players (i.e. buyers, suppliers, competitors and complementers) are shifting. In these markets, dynamic capabilities necessarily rely much less on existing knowledge and much more on rapidly creating situation-specific new knowledge (Eisenhardt & Martin, 2000). In high velocity markets, dynamic capabilities are simple, experiential, unstable processes relying on quickly created knowledge & iterative execution to produce adaptive but unpredictable outcomes.

(24)

With the addition of the concept of market dynamism, the concept of dynamic capabilities is further integrated with the environment of organizations. It is thus important for managers to estimate the market velocity.

Maturity knowledge domains

The second dimension of the knowledge strategy framework is the maturity of knowledge domains. Van Agteren (2004) takes an evolutionary approach to knowledge domains as he makes a distinction between young and mature knowledge domains. He advocates that there are important differences between the two. First, it is assumed that when a new knowledge domain is established it will have few experts and organizational links. It takes time to build knowledge about a specific topic in an organization. In an evolutionary perspective it seems more appropriate to see the development of knowledge domains as a continuum ranging starting as a young knowledge domain and evolving to a mature knowledge domain inside the organization. This makes the determination of the maturity of knowledge domains more difficult, but also more realistic. Second, he assumes that the qualities of knowledge in a young versus mature knowledge domain differ. One difference between knowledge might be the amount of explicit and tacit knowledge available to the knowledge domain. Because more knowledge must be created in a young knowledge domain than in a mature knowledge domain it is assumed that young knowledge domains consist of more tacit knowledge. The same can said about the degree of integration of a knowledge domain with the rest of the organization. Because young knowledge domains are not well established it is assumed that the links with the rest of the organization are weak in comparison with mature knowledge domains. The disadvantage of this might be that young knowledge domains work less efficient than mature knowledge domains, but the advantage may be that young knowledge domains are more flexible. The maturity of a knowledge domain can be established by investigating the maturity and quality of its knowledge elements and Community of Practice. In the next sections the dynamic capability types and the four knowledge strategies will be discussed.

Dynamic capability types

The knowledge strategy framework assumes that in each configuration of the model knowledge is differently modified. In order identify how knowledge is modified in each knowledge strategy, first the ways in which knowledge can be modified will be discussed. Eisenhardt & Martin (2000) have identified four types of dynamic capabilities, which are integration, reconfiguring, gaining, and releasing dynamic capabilities.

Dynamic capabilities are ways in which knowledge can be modified. Integration dynamic capabilities are product development routines by which managers combine their varied skills and functional backgrounds to create revenue-producing products and services. A typical aspect of product development is that it involves the combination of knowledge from multiple functional backgrounds. Product development could also consist of the combination of multiple market backgrounds. Reconfiguration dynamic capabilities are according to Eisenhardt & Martin (2000) transfer processes including routines for replication and brokering which are used by managers to copy, transfer and recombine resources, especially knowledge-based ones, within the firm. With regard to knowledge routines a clear example of reconfiguration is that managers routinely create new products by knowledge brokering from a variety of previous design projects in many industries and from many clients.

(25)

Gaining dynamic capabilities includes knowledge creation routines whereby managers and others build new thinking within the firm, a particularly crucial dynamic capability in industries like pharmaceuticals, optical disks, and oil where cutting-edge knowledge is essential for effective strategy and performance. They also include alliance and acquisition routines that bring new resources into the firm from external sources. Biotech firms with strong alliancing processes for accessing outside knowledge achieve superior performance (Powell et al., 1996). Finally, releasing dynamic capabilities are, although often neglected, exit routines that jettison resource combinations that no longer provides competitive advantage. Releasing dynamic capabilities are also critical dynamic capabilities as markets undergo change (Sull, 1999a, 1999b).

Types of knowledge strategies

Here the four identified knowledge strategies will be elaborated. When the two dimensions of the knowledge strategy framework are combined four knowledge domain configurations can be identified. In each configuration knowledge is modified by different dynamic capabilities (see figure 3.6).

Applying strategy: An applying strategy is most effective in a mature knowledge domain and in a moderately dynamic environment. In this strategy the dominant dynamic capability is reconfiguring of knowledge over integrating, gaining and releasing dynamic capabilities. The reason for this is that the knowledge in this knowledge domain degenerates slowly and can be reapplied.

Appropriating strategy: An appropriation strategy is most effective in a young knowledge domain and in a moderately dynamic environment. In this strategy the dominant dynamic capabilities are reconfiguring and gaining of knowledge over integrating, and releasing dynamic capabilities. The reason for this is that in a young knowledge domain new knowledge must be gained, which degenerates slowly.

Expanding strategy: An expanding strategy is most effective in a mature knowledge domain and in a high velocity dynamic environment. In this strategy the dominant dynamic capabilities are reconfiguring, gaining and releasing of knowledge over integrating dynamic capabilities. The reason for this is that existing knowledge must be released and new knowledge must be gained in order to match the rapidly changing environment.

Innovating strategy: An innovating strategy is most effective in a young knowledge domain and in a high velocity dynamic environment. In this strategy the dominant dynamic capabilities are integrating, and gaining of knowledge over reconfiguring, and releasing dynamic capabilities. The reason for this is that the knowledge from multiple sources must be integrated and gained quickly before the knowledge becomes obsolete.

(26)

Conclusion

After the identification that the unscheduled down process is problematic in the diagnostic problem statement, the multiform analysis identified and discussed two promising theoretical perspectives. These perspectives are an organisational and a knowledge management perspective and will be used for diagnosing and solving the organisational problem. First, an organisational perspective discussed theories which are felt to be important. These theories are criteria for organisational performance, a discussion of structured and unstructured problems, organisational structure, coordination of interdependences, and organisational boundaries. Secondly, the knowledge management perspective set the field for investigating organisational problem from a knowledge management perspective. It discussed various subjects, such as knowledge, knowledge elements, local and organisational learning, knowledge management, and a knowledge strategy framework. In the remaining part of this research the theoretical perspectives will be used in order to further diagnose and improve the unscheduled down process at ASML.

(27)

3.3 EVALUATION Introduction

According to De Leeuw (1996) the central question in the evaluation of an organizational problem is ‘who is worried about what exactly?’. This is a very important question since it identifies and it defines the organizational problem. In answering this question the following sub-questions must be answered. First will be investigated if the organisational problem is a functional problem in the sense that it affects the functioning of the organisation. After this a so-called problem owner analysis is performed in order pinpoint the subjective discomfort of the actors in the unscheduled down process. From the insights of the problem owner analysis it is possible to distil the reality problem. When the reality problem is solved it is believed that the majority of the subjective problems of the problem owners are solved. After this the system boundary for solving the organizational problem can be constructed which defines the organizational problem exactly. After this it will be evaluated to which degree the organisational problem is problematic. The final section investigates if there are differences between the formal, and actual organization. If there are huge differences the effectiveness of formal measures is limited which is an important insight for the diagnosis of the organizational problem.

Functional problem

Before will be started with the evaluation of the organisational problem, it must be remarked that the evaluation of the organisation problem must be focused on so-called functional problems (De Leeuw, 1996). Functional problems are problems which have a direct relationship with the functioning of the organisation. For example, the functioning of a car is determined by for example its fuel usage, and acceleration. In the case of the ASML machines its functioning is affected by if it operates, if the machine performs according to specifications, if the output is sufficient, and so forth. The problem in this research is the unscheduled down process, and since this focuses on if the machines functions as a whole, this is indeed a functional problem.

Problem owner analysis

De Leeuw (1997) defines a problem as a subjective discomfort of a problem owner mixed with the desire to resolve this discomfort. In order to solve an organisational problem this subjective discomfort must be understood. This is done by a problem owner analysis. The analysis is based on a ten-day local research in Dresden which is described in Appendix A. The organisational problem of this research is that the unscheduled down process must be improved. In this process multiple actors at multiple levels are involved. In this research four different types of actors can be identified: the customer, the field service engineer (FSE), the regional office (with 2^nd line engineers, performance co-ordinators and customer support supervisor), and the headquarters (with 3^rd line engineers and supporting departments). The involved actors and their relationships are graphically depicted in Appendix C. Each actor has its own perspective in the way in which unscheduled downs are solved and why this may be problematic.

Customer:

The customer is worried about the way in which unscheduled downs are processed, because the machines are down too often and for too long. The uptime of the machines is seen as insufficient, which results in loss of production.

(28)

The customer believes that the reason for the many and long unscheduled downs are that the diagnosis process by FSE’s is insufficient, and that not enough spare parts are on stock. To the customers opinion the work on the machines is done with too many people and this should be reduced by half in the future.

ASML FSE’s:

The FSE’s are worried about the process of unscheduled downs, because their workload is too high and because they receive negative feedback from the customer and regional office about this process. The FSE’s explanation for the long downs is that there are not enough engineers available for solving unscheduled downs. Furthermore, at night and in weekends the process of unscheduled downs is negatively affected by the lack of support from the 2^nd lines, and the headquarters in Veldhoven.

ASML regional office Dresden:

The regional office is worried about the process of unscheduled downs, because the workload for the regional office is too high, performance targets are not met, and the customer is unsatisfied. The explanation for this situation is that to there are not enough FSE’s and 2^nd line engineers which leads to a to high workload. This is especially true when multiple machines are down at the same time.

ASML headquarters:

The ASML headquarters exists of two groups: the third line engineers and the supporting departments. The third line engineers are worried about the unscheduled down process, because they receive many escalations from regional offices. The third line engineers feel that the unscheduled down process could be improved if the FSE’s and the second line engineers would follow procedures. The support departments have the responsibility to support and improve the unscheduled down process. These departments are worried about the way in which unscheduled downs are processed, because this leads to not meeting their support targets. These departments feel that too many unscheduled downs are escalated to the 3^rd line engineers in the headquarters, which could be solved at the regional office if escalation procedures would have been followed.

Reality problem

Based on the problem owner analysis it will be attempted to construct the reality problem of this investigation. It is important to construct the reality problem of this research out of the subjective problems in order to compose the objective organisational problem.

According to De Leeuw (1997) a reality problem is the for problem solution relevant concrete system like it is defined by the researcher. The problem owner analysis shows that the current situation with unscheduled downs has negative consequences for all involved parties. The reason why an actor is anxious differs depending on his role in the process and his responsibilities. The FSE’s and the regional office, for example, think that the low man-machine ratio is an important explanation for the current unscheduled down process. Since the man-machine ratio dropped from 1:3 to 1:9 in a period of two years, the workload for FSE’s has increased substantially. But since the man-machine ratio cannot be influenced in the short run and it does not give insight into the quality of the current unscheduled down-solving process, this will be out of the scope of this research.

(29)

The result of the low man-machine ratio is that the workload for the FSE’s and the regional office has become too high and that their activities now mainly exist from ad hoc problem solving (‘fire fighting’). In ad hoc problem solving there is no time left for supporting activities such as administration, analysis and reporting of unscheduled downs, and preventive maintenance. The FSE’s complained that at nights and in the weekends there is not enough support, points that the FSE’s are highly dependent of 2^nd and 3^rd line support. The decrease of the man-machine ratio requires a different way of working. The fact that the same job has to be done by fewer people makes it imperative to use available knowledge as much as possible.

From this analysis it is believed that optimizing the use of knowledge elements which support the unscheduled down process solves the majority of the problems of the problem owners. This becomes clear by describing the expected effect of improving the use of knowledge elements of the problem owners.

Customer: if available world-wide knowledge in the shape of knowledge elements are optimally used, it is expected that the unscheduled down time of the machines shall decrease and the uptime of the machines will be optimized.

ASML FSE’s: if the FSE’s use available knowledge elements optimally, it is expected that their workload shall decrease and that they will receive less negative feedback from the customer and the regional office.

Regional office: if knowledge elements are optimally used, it is predicted that the procedures will be followed, which shall result in a decrease of the regional office workload.

Headquarters: when knowledge elements are optimally used, it is expected that the workload at the headquarters shall decrease, and that the targets of the supporting departments will be met.

Thus the not optimal use of knowledge elements is the reality problem of this research.

In an attempt to narrow down the problem as much as possible it will be investigated here why exactly the knowledge elements are not used. Since the engineers are professionals with a high degree of independence they shall only use knowledge elements when they are willing and capable of using them. Possible explanations of why these elements may not be used can be seen in table 3.3.

Not capable Insufficient quality of knowledge elements

Lack of skills

No time

Insufficient resources such as information, laptop, tools

Insufficient support from regional office and headquarters Not willing Not managed (rewards or penalties)

Culture: knowledge elements do not need to be used

No priority

Table 3.4 Possible reasons for not using knowledge elements

Being capable and willing to perform procedures seem to be very well related as can be seen in figure 3.7. When a procedure is not functioning well (being capable), it is unlikely that the user is willing to use it a second time. When an engineer is not willing to use a software tool and he has never done this, he may not be capable to use this tool.

A practical knowledge management approach

Effectiveness through managing knowledge