ASTER T HESIS M

(1)

M

ASTER

T

HESIS

Public Version

T

OWARDS IMPROVING MAINTENANCE PERFORMANCE

(2)

A

BSTRACT

In this study, a framework to measure maintenance performance in a military context is proposed. Moreover, a method for the implementation of dynamic maintenance will be presented. This is a step-by-step plan which can be used for the implementation of dynamic maintenance Relevant literature argues that implementing dynamic maintenance will result in an improvement of the cost-efficiency of platforms (Wubben, 2009). However, there is no elaboration on these concepts and it is not known what the benefits of such a concept are. Therefore, a case study has been conducted at the Netherlands Ministry of Defence to be able to implement and test the proposed concepts. The most important argument made is that enhancing a dynamic maintenance concept will result in a better matching of actual usage and the actual maintenance demand of a complex system. This results in an increase in the maintenance performance.

Keywords;

Maintenance performance, Maintenance strategies, dynamic maintenance concept, implementing dynamic maintenance, usage load based maintenance, ULBM, usage severity based maintenance,

USBM, CV90, degrader analysis, Netherlands Ministry of Defence, usage profiles.

(3)

P

REFACE

After nearly six months, more than 10,000 kilometers in public transport, many interviews and meetings all over the country, many days of writing, inspirational runs on the beach of Den Helder and a lot of coffee, this report ends a very interesting period of study at the Dutch Ministry of Defence. I’ve not only learned a lot about maintenance, but also about the course of business at this ministry. This study has enabled me to take a close look at different maintenance departments, the Defence Materiel Organization, operational units and many supporting organizations. As I have already been working for the Dutch army reserve, a lot of procedures and practices were already familiar to me. However, conducting this study gave me a great look behind the scenes and the opportunity to learn a lot about the organization.

This study could not have been what it is without the support of my supervisors Mr. Tinga of the Netherlands Defence Academy and Mr. Wubben of the University of Groningen and Netherlands Defence Academy. I hereby want to thank them for the very good feedback and sharp comments. I liked the way they were thinking along with me and proposed interesting ideas or reports and motivated me to make the best of this study. In hindsight, I can say that they might have saved me a couple of times from going the wrong direction and helped me to solve some important issues. Especially in the final weeks they helped me to solve important issues. Next to them, I’d like to thank professor Vis for her sharp comments which changed this report radically.

Furthermore, I owe gratitude to the many people in the organization who helped me conducting my study. I will not call them by name but I’ve been positively surprised by many people who took a lot of time for me and sometimes even gave me great feedback on my report. Moreover, they’ve kept me sharp in meetings and conversations by criticizing the concepts, my study, or my (sometimes too) enthusiastic ideas.

Finally, I hope that this study will contribute to the tile construction of dynamic maintenance reports of the University of Groningen and the Netherlands Defence Academy and an improvement of the maintenance process of the Dutch Ministry of Defence.

(4)

APPENDIX 4A:MAINTENANCE STRATEGY ... FOUT!BLADWIJZER NIET GEDEFINIEERD. Design of the maintenance function ... Fout! Bladwijzer niet gedefinieerd. maintenance Policies ... Fout! Bladwijzer niet gedefinieerd. Further specification of the maintenance strategy ... Fout! Bladwijzer niet gedefinieerd. APPENDIX 4B:THEORETICAL AVAILABILITY ... FOUT!BLADWIJZER NIET GEDEFINIEERD. APPENDIX 4C:VEHICLE INFORMATION SYSTEM DIS/VIS... FOUT!BLADWIJZER NIET GEDEFINIEERD. APPENDIX 5: MAINTENANCE POLICIES DESCRIPTION ... FOUT! BLADWIJZER NIET GEDEFINIEERD. APPENDIX 6: VEHICLE LOGS DESCRIPTION ... FOUT! BLADWIJZER NIET GEDEFINIEERD. APPENDIX 7: STANAG 2985 ... - 7 -

APPENDIX 8: DESCRIPTION OF CRITICAL COMPONENTS ... - 8 -

APPENDIX 9: ULBM FACTORS ... - 10 -

APPENDIX 10: CALCULATION OF FAILURES OF TRACK PAD ... - 14 -

(6)

G

LOSSARY OF TERMS

Abbreviation In English In Dutch

CBM Condition Based Maintenance Toestand Afhankelijk Onderhoud

CLAS Dutch Army Commando Land Strijdkrachten

CV90 Combat Vehicle 90 Combat Vehicle 90

AGB Defense Estate Company Algemeen Goederen Bedrijf

DIS Diagnostic Information System Diagnostisch Informatie Systeem DMO Defence materiel organization Defensie Materieel Organisatie LBM Load Based Maintenance

MTBF Mean Time Between Failure MTTR Mean Time To Repair

NLDA Netherlands Defence Academy Nederlandse Defensie Academie NL MOD Netherlands Ministry of Defence Nederlands Ministerie van Defensie

TBM Time Based Maintenance Tijdsafhankelijk onderhoud

(7)

1. I

NTRODUCTION

Recent budget cuts on defense departments and a rise in various deployments of military units all over the world have created an increased pressure on these military organizations. High availability of weapon systems has to be ensured by conducting maintenance (Moubray, 1997). However, it has been found that the availability of military systems is structural lower than comparable civil systems. This is explained by Werkman (2009) who argues that military systems are deployed in extreme harsh environments (e.g. sandy regions like Afghanistan and Iraq) and are exposed to heavy loads (e.g. ballistic protection on trucks). This forms an important problem for defense organizations since dramatically high costs are associated with ensuring availability of these, often very expensive, systems. Furthermore, Werkman (2009), who studied the availability of weapon systems of the Dutch Army in Afghanistan, concludes that the maintenance department is often not prepared and tailored to these situations. Wubben (2009) argues that by implementing a dynamic maintenance strategy, the cost-effectiveness of a weapon system can be improved because the maintenance organization will be tailored to the actual maintenance need of a platform. However, as Simões, Gomes and Yasin (2011) argue, it is essential to approach maintenance management systematically and strategically to make the right choices, especially in capital-intensive industries. Moreover, the defense organizations have to be aware of the factors that influence the outcome of their actions to be able to improve these. Kennerly and Neely (2003) and Lingle and Schiemann (1996) have observed that companies which use a performance measurement system perform better than those who do not measure performance. Moreover, companies can use performance measurement to monitor implementation of new plans and determine if they are successful and how they can be improved (Atkinson, Waterhouse, Wells, 1997). Therefore, a performance measurement system should be used to assess the maintenance function and to concretize the effects of improvements. The first objective of this study therefore is: “To design a framework capable of measuring maintenance performance in a military context”.

1.1 R

ELEVANCE

The maintenance department in a military environment differs significantly from a civil (private) maintenance department. Doerr, Eaton and Lewis (2004) point out the measurement issues in performance based logistics in a military context. They argue that for a private-sector vendor, the primary objective is to maximize wealth. However, the objective of the user (the department of defense) is harder to assess. Doer et al (2004) argue it is: “to gain more security for the nation”. This creates a difficult translation problem as it is not clear how to measure the value of the service provided to the department of defense in terms of dollars, except for the price they pay for it. This yields also for the performance delivered by the (internal) military maintenance organizations. It is not clear how to evaluate the value of extra availability or higher reliability of a system. Next to that, maintenance crews in military organizations have to be trained to ensure their military skills and drills because these crews join operational units on their deployments or multiple-day training exercises. Furthermore, the varying deployments in often hostile environments require a complex supporting organization to be able to repair defect systems during these deployments.

(8)

dynamic maintenance strategies for military systems (e.g. Wubben, 2009). These dynamic strategies tailor the maintenance needs to the actual demands and improve thereby the cost-effectiveness of the system.

However, there is no elaboration on these dynamic maintenance concepts and it does not become clear how these systems should be implemented and what (if any) the advantages and differences are compared to static maintenance strategies. By comparing the arguments and observations made by Werkman (2009) and Wubben (2009), the following hypothesis, which will be tested in this study, has been formulated: “By introducing a dynamic maintenance strategy, the maintenance performance will increase”. Since it is not described in relevant literature how dynamic maintenance should be implemented, the second research objective in this study is: “To design a method for the implementation of a dynamic maintenance concept”.

This study will attempt to cover these white spots in literature. First, a framework will be proposed to measure the maintenance performance in a military context. Moreover, this framework should be capable of concretizing improvements made in the maintenance department. Second, since relevant literature argues that the performance could be improved by enhancing dynamic maintenance strategies, but it has not been made clear how these should be implemented, a method for the implementation of dynamic maintenance will be presented. To be able to validate these frameworks, test the hypothesis and visualize the effects of the implementation of a dynamic maintenance strategy, a case study of a military vehicle (CV9035NL) at the Dutch Ministry of Defense will be conducted.

1.2 R

ESEARCH QUESTION

The both articulated objectives can be combined in the following research objective:

“To design a framework capable of measuring maintenance performance in a military context and to design a method for the implementation of dynamic maintenance”.

Following the research objective, the research question has been formulated as:

“How could the maintenance performance in a military context be measured and improved by implementing dynamic maintenance?”

To be able to accomplish the research objective, test the hypothesis and answer the research question. The following sub-questions are formulated. To be able to validate the hypothesis and study how dynamic maintenance affects the maintenance performance a question has been formulated. This question (7) can therefore be classified as a case study related question.

The sub-questions are:

1. What are the factors that influence maintenance performance?

2. What frameworks for the measurement of maintenance can be used as a basis for the introduction of a framework for measuring maintenance performance in a military context? 3. How should the maintenance performance measurement be measured in a military context? 4. What is dynamic maintenance?

5. What methods for the implementation of dynamic maintenance could be used as a basis for the implementation of dynamic maintenance?

6. How should dynamic maintenance be implemented?

(9)

1.3 S

COPE AND LIMITATIONS

The study will focus on the measurement of maintenance performance in a military context. One way to improve this performance is, according to relevant literature, the implementation of dynamic maintenance. Therefore, this study will focus on the (effects of) implementation of dynamic maintenance as a method to improve the maintenance performance. It is thereby limited by not regarding other methods to improve this performance. Furthermore, since the focus is on a military context, measurement and implementation issues in other sectors will not be included directly.

1.4 S

CIENTIFIC RELEVANCE

By conducting this study, scientific relevance will be gathered from two points:

 The introduction of a framework for measuring the maintenance performance in a military context. Frameworks for maintenance performance are designed by several authors. However, in a military context, where systems are not used 24/7 and the value of availability during a training or deployment period is hard, if even, to measure in Euros or Dollars, these existing frameworks are not sufficient.

 Furthermore, this study will show how a dynamic maintenance strategy can be implemented and will visualize the effects of implementation. Moreover, in the case study an example of dynamic maintenance profiles will be created for a military system. This has never been done in relevant literature.

1.5 L

INK BETWEEN SUB

-QUESTIONS AND CHAPTERS

Questions 1, 2, 4 and 5 are theory-related questions. These will be answered in the literature study (Chapter 3). Questions 3 and 6 will be answered in the framework development (Chapter 4). Question 7 will be answered in the case study (Chapter 5). The validation of questions 3 and 6, using question 7, will be discussed in Chapter 6.

1.6 T

HESIS STRUCTURE

(10)

2. R

ESEARCH

M

ETHODOLOGY

The research methodology forms the basis of this study. In this chapter will be explained how this study will be conducted. As has been argued in the introduction, relevant literature describes many frameworks to assess the maintenance performance. However, there is no maintenance performance framework suitable to use in a military context. Moreover, a framework to implement a dynamic maintenance strategy is lacking. This chapter will elaborate on the methodology to design and test these frameworks. Moreover, the methodology to conduct a well-founded case study will be discussed.

2.1 R

ESEARCH APPROACH

The purpose of this study is to develop a framework for measuring maintenance performance and a method to implement dynamic maintenance. Since the proposed frameworks are new, they have to be tested using a case study. Therefore, an empirical research approach is used to be able to test the findings in a practical situation. Moreover, induction of the finding will be applied to generalize conclusions made from studying one specific case.

O’Leary (2010) argues that two traditions in conducting research can be identified. Firstly, there is the quantitative tradition and secondly there is the qualitative tradition. Quantitative research is often characterized as an objective positivist search for singular truths that relies on hypotheses, variables and statistics, is generally large scale, without much depth (O’Leary, 2010). While qualitative research studies, reject positivist rules and work at accepting multiple realities through the study of a small number of in-depth cases (O’Leary, 2010). This study could be regarded as qualitative study. It will use small scale interviewing, expert reviews, observations, document analysis and relevant literature. Quantitative data, gathered in the case study, will be used to support the qualitative research.

Qualitative studies call on inductive as well as deductive logic, appreciate subjectivities, accept multiple perspectives and realities, recognize the power of research on both participants and researchers, and do not necessarily shy away from political agendas (O’Leary, 2010). Moreover, a qualitative research project delves into the (social) complexities of the analyzed system. This will, especially in the ‘Implementation’ phase, be the case in this study. Understanding of the complexities helps to fully understand the social system and helps to improve the recommendations of this study. This study is build up as following. First, by studying relevant literature and using preliminary observations, a number of white spots are found in literature. This is briefly described in the introduction. In the literature study (Chapter 3) this will be elaborated on. To be able to cover these white spots, two frameworks will be proposed. These frameworks will be tested using a case study. Finally, the sections can be combined to be able to discuss the differences and draw conclusions.

2.2 C

ONDUCTING THE

CV90 CASE S TUDY

To be able to verify the proposed framework and presented method for implementing dynamic maintenance, a case will be selected to demonstrate the value of both frameworks. This process of testing the frameworks has many similarities with a case study. Therefore, the guidelines for conducting a case study will be followed. Yin (1989) defines a number of characteristics of a case study:

1. The type of research question: typically to answer questions like “how” or “why”;

(11)

2. Extent of control over behavioral events: when investigator has a little/no possibility to control the events;

The case study serves mainly as a source to gather data from. Aspects like the current state of the system, costs of maintaining the system cannot be influenced by the researcher. In such way, there is no action-research involved in this study. However, the creativity of the author will be used to improve the performance and the outcome relies partly on cooperation within the company. Therefore, the researcher should be aware of these effects.

3. General circumstances of the phenomenon to be studied: contemporary phenomenon in a real-life context.

The case study is about a phenomenon in real-life.

From this list of characteristics can be concluded that the case-study concept can be used for this study. To be able to conduct a well-founded case study, the five steps of conducting a case study, defined by Stuart et al (2008) can be followed. These steps, as visualized in Figure 1 should be used to design this case study.

FIGURE 1,FIVE-STAGE RESEARCH MODEL (STUART ET AL.,2002)

Stuart et al (2002) note that the development of a case study entails much more than organizing the questions the researchers are going to pose. Therefore, it will be discussed below how these five steps are implemented in this study.

Define the research question

By defining the research question, the focus of the study is clearly given. This is especially important in exploratory studies, because the findings could be hard to predict. Moreover, since the results are unknown of forehand, a more interesting research question could be defined during the study and thereby a shift in the focus of the study. However, if this might happen, the entire protocol of the study needs rework (Stuart et al., 2002). The research question used in this study is given in the introduction.

Instrument development

To be able to verify the frameworks, an instrument should be developed which should make clear how data gathered in the case study will be linked to the final conclusions (Yin, 1989). In this study, a bottom-up approach will be used. According to Yin (1989), constructing validity and internal validity is the main concern of a case study. This means that the measurements should prove that conjectured relationships really exist and reflect the phenomena studied. This is taken care by, through creating a number of frameworks and comparing these to relevant literature and the CV90 case. Moreover, the hypothesis which will be tested can be evaluated with calculations made on the CV90 case. Therefore, the instrument consists of a questionnaire, used to gather the data, and the proposed frameworks. These should be used to guide the process and, together with the gathered data, conclusions can be drawn.

Data Gathering

(12)

will be explained in the case study), the most recent data available to the researcher will be used. Moreover, as the CV90 is only three years in use, taking into account a phased introduction, not all CV90s are already in use for three years, this reduces the amount of available data significantly. Next to the gathering of ‘hard data’ interviews will be held with experts and users. Structured surveys held as face-to-face interviews will be used to generate valid data. These surveys are included in Appendix 1, Surveys. Data gathered from documents will be checked and discussed in expert interviews. This triangulation of data will help to ensure the validity of the obtained data. Since a case study uses theoretical and analytical generalization of data, logical extrapolation will be conducted (Cf. Stuart et al, 2002). Moreover, the researcher should judge if a particular finding can be generalized to be able to use in other circumstances (Stuart et al, 2002). Finally, the researchers should be aware that other factors which aren’t measured or taken into account could give another explanation of the patterns observed (Stuart et al, 2002).

Analyze Data

After the data has been obtained, it should be analyzed carefully to generalize conclusions. Since only one case study will be conducted, drawing conclusions should be done extra carefully. Moreover, this means that the researcher cannot claim truth. Therefore, the conclusions could better be labelled as ‘propositions’. These propositions should be tested against larger data samples to draw conclusions. Disseminate

The results of the case study should be presented in a credible way to avoid criticism on the study. Therefore, the concerns made above should be kept in mind.

2.3 C

ONCLUSION OF METHODOLOGY

(13)

3. L

ITERATURE REVIEW

In this section, the sub-questions 1: What are the factors that influence maintenance performance?, 2: What frameworks for the measurement of maintenance can be used as a basis for the introduction of a framework for measuring maintenance performance in a military context?, 4: What is dynamic maintenance? and 5 : What methods for the implementation of dynamic maintenance could be used as a basis for the implementation of dynamic maintenance?, will be answered using perspectives discussed in relevant literature. Therefore, first the need of maintenance will be explained, this will be followed by giving an overview of different maintenance concepts and strategies. In section 3.3, available frameworks for the measurement and improvement of maintenance performance will be discussed. Section 3.4 discusses dynamic maintenance concepts. In section 3.5, the design criteria mentioned in this chapter for the creation of a framework for measuring maintenance performance and a method for the implementation of a dynamic maintenance concept will be listed. This section will end with a brief conclusion where the discussed sub-questions will be answered; this section also serves as input for the following chapters.

To prevail in battle, the operational availability of platforms is of high importance. Operational availability (A0) is a function of reliability, maintainability and supportability (Manary, Price and Weinstein, 2003). It measures the average availability of a platform and includes all sources of downtime; it is measured by dividing the mean time between failures by the mean time to failure plus the mean time to repair and the mean logistics downtime.

However, to be able to use a platform, as will be shown in Figure 2, not only the availability of the technical (physical) part of the weapon system is required, but also consumables like fuel and batteries as well as operators to control and use the system should be available. By noting this, it can be concluded that the aggregation level of which the platform is regarded should be determined. Regarding it from a very high aggregation level, operators and consumables have to be included because these are needed to deploy the platform in for example a conflict situation (the need for an deployable platform). Regarding it from a lower level, parts of the system could be viewed as separate systems and these systems can be studied individually.

Available platform Maintainability / Serviceability Consumables Operators Deployable platform Reliability

FIGURE 2,INGREDIENTS NEEDED FO R A DEPLOYABLE PLATFORM

(14)

3.1 M

AINTENANCE

The British Standards Institute (1984) defines maintenance as: “A combination of any actions carried out to retain an item in, or restore it to an acceptable condition”. This acceptable condition is a very relative term. However, when a combat vehicle is bought, it has been specified that it should for example be able to drive 70 km/h and shoot and hit all targets with a fire rate of 300 shots / minute. This performance should ideally be available at all time; this means an operational availability of 100%. Moreover, system breakdowns should not occur when the system is in use. Therefore, the maintenance function of the company should retain the items to an acceptable condition (cf. The British Standards Institute, 1984). The extent to which the system should be able to perform the intended functions (the acceptable condition) has to be defined by the user. Moreover, since budgets are limited, maintenance should be reduced to a minimum. This will lead to a trade-off between the extent to which the system is able to perform the intended functions and the costs associated with achieving this. Presumably, when more maintenance is carried out, the system will, to a certain extent, be better able to perform the intended functions.

Moubray (1997) defines maintenance as: “Ensuring that physical assets continue to do what their users want them to do.” The technology of maintenance includes finding and applying suitable ways of managing techniques that include predictive and preventive maintenance, failure-finding and run-to-failure (Moubray, 2000). Tsang (2002) identifies four strategic dimensions of maintenance, given in Figure 3. Dimension Service-delivery options Organization and work structuring Maintenance methodology Support systems Description: The choice between

in-house capability and outsourced service

Organization of the maintenance function and the way maintenance tasks are structured

The selection of maintenance policies Design of the infrastructure that supports maintenance

Strategic options: Focus on

maintenance as a core competency Outsource maintenance activities which are not part of the company’s core competencies Flatten the hierarchy Develop a flexible workforce Maintain a specialized workforce Focus on asset-centered methodology – RCM Focus on people centered methodology – TPM Empower the employees Cultivate teamwork

FIGURE 3,DIMENSIONS OF MAINTENANCE (TSANG,2002)

The selected mix of strategic options of the four dimensions of maintenance described by Tsang (2002) forms an approach for the design of the maintenance function. The design of the maintenance function of a firm is described as the maintenance strategy (for example: Kevin and Penlesky, 1988; Pintelon and Pinjala, 2006; Pinjala and Pintelon, 2004).

3.2 M

AINTENANCE STRATEGIES

(15)

different strategies described in relevant literature. Firstly, a definition of maintenance strategies will be given. Secondly, the distinction between the different maintenance policies will be discussed.

3.2.1 D

EFINING M AINTENANCE STRATEGIE S

Pintelon and Pinjana (2006) argue that the definition of maintenance strategies in relevant literature is either too narrow or too vague. For example, Swanson (2001) distinguishes three maintenance strategies: reactive, proactive, and aggressive maintenance. Next to that, maintenance policies like preventive maintenance, corrective maintenance, and predictive maintenance or maintenance concepts like reliability-centered maintenance (RCM) and total productive maintenance (TPM) are often seen as maintenance strategy (see for example: Bevilacqua and Braglia, 2000; Moubray, 1997). Dekker (1996) and Takata et al (2004) argue that the maintenance concept or strategy describes what events (e.g., failure, passing of time) trigger what type of maintenance (inspection repair, replacement). This creates the difference in regarding the maintenance strategy as a dimension of maintenance, or as seeing it as the total design of the maintenance functions. The latter is done by Kevin and Penlesky (1988). They regard maintenance strategy as: a mix of elements like maintenance policies, backup equipment and equipment upgrades. This is complemented by Kelly (1997). He regards a maintenance strategy as the identification, resource allocation and execution of repair, inspection and replacement decisions. Tsang (1998) sees a maintenance strategy as for example asset utilization, improving responsiveness or a focus on developing core competencies.

Pintelon and Pinjana (2006) argue that maintenance should be seen in a broad perspective. A maintenance strategy is deﬁned as a series of uniﬁed and integrated pattern of decisions made in four structural and six infrastructure decision elements (Pinjala and Pintelon, 2004). These decision elements are visualized Table 1. In this definition a maintenance strategy is regarded at a functional hierarchy level, similar to manufacturing or any other function (Pinjala and Pintelon, 2006).

Structural decision elements Infrastructure decision elements

Maintenance capacity: work force, supervisory, and

management staff

Maintenance organization: organizational structure Maintenance facility: Tools, equipment, spares,

workforce specialization, location of workforce

Maintenance policy and concept: corrective,

preventive, predictive maintenance. Concepts like TPM, RCM.

Maintenance technology: predictive maintenance, or

condition monitoring technology, expert system, maintenance technology

Maintenance planning and control system:

maintenance activity planning, scheduling, control of spares, costs, etc.

Vertical integration: in-house maintenance versus

outsourcing

Human resources: recruitment policies, training and

development of workforce and staff, culture and management style

Maintenance modifications: equipment design

improvements, new equipment installations, new machine design support

Maintenance performance measurement and reward systems: performance recognition, overall

equipment effectiveness, balanced scorecard

TABLE 1, MAINTENANCE STRATEGY DECISION ELEMENTS (PINTELON AND PINJALA,2006)

(16)

Criteria for providing treatment - Time

- Condition - Detection of breakdown - Detection of symptom - Analysis of trend Opportunity of maintenance task execution

(inspection/monitoring/diagnosis/treatment)

- During operation - During stoppage - When disassembled

Type of treatment - Servicing

- Repair - Replacement

- Design improvement

TABLE 2,REASONS FOR MAINTENANCE POLICIES (TAKATA ET AL.,2004)

Takata et al (2004) propose a model for the selection of a maintenance strategy. They mention five factors which should be evaluated to choose the right strategy. The difference between environmental conditions (temperature, weather, pressure, and other variables impacting the deterioration of the system) and operational conditions (what the user expects from the system) in their model is that the operational factors could be controlled by the user whereas the environmental conditions cannot (Cocheteux, Voisin, Levrat, and Iung, 2010). The model of Takata et al (2004) is given in Figure 4. Below, the terms in this model are explained briefly.

FIGURE 4,FACTORS DETERMINING MAINTENANCE STRATEGI ES (TAKATA ET AL.,2004)

 Managerial features of facilities: This factor influences the features of the equipment. It includes planning, capability and quality of personnel and maintenance departments.

 Operational conditions: What the user expects from the system. This will be influenced by the demand of the users of the equipment.

 Environmental conditions: temperature, weather, pressure, and other variables impacting the deterioration of the equipment.

 Available maintenance technologies: This factor influences the applicability of maintenance technologies. For example, the availability of an AIDA system (which measures the deterioration of the system).

(17)

 Managerial aspects: To evaluate the managerial aspects, objects to be maintained should be prioritized for the purpose of resource allocation.

 Technological aspects: To evaluate the technical aspects, maintenance strategies which are technical applicable should be selected.

Used definition of maintenance strategy:

In this study, the view of Pintelon and Pinjana (2006) is enhanced. A maintenance strategy should not be seen as just a maintenance policy. The definitions of Kevin and Penlesky (1988) and Kelly (1997) fit the study best and are therefore combined and used in this study.

A maintenance strategy is a mix of maintenance policies, backup equipment, equipment upgrades, and the identification, resource allocation and execution of repair, inspection and replacement

decisions.

This chosen maintenance strategy should not per definition be enhanced for the whole maintained system. Banks et al (2008) argue that a combination of strategies provides an optimal maintenance solution for a complex system. Furthermore, they state that a maintenance plan that minimizes the amount of reactive maintenance and appropriately uses preventive maintenance and condition based maintenance methodologies where they will be most effective, results in an optimum maintenance approach for a given system.

3.2.2 M

AINTENANCE POLICIES

The maintenance strategy is a set of decisions which influence the total maintenance process. One important factor of the maintenance strategy is the maintenance policy (however, many authors call this the strategy). This policy influences the moment of conducting maintenance and the type of maintenance which will be conducted. Using the relevant maintenance literature described in chapter 3.1, a distinction between the various maintenance policies will be proposed; these are explained below and visualized in Figure 5.

(18)

Maintenance

Corrective Detective

Reactive Proactive Aggressive

Preventive Predictive Time-driven Usage-based Design improvement Opportunistic Dynamic Condition-based Scheduled Usage

severity based Load based

FIGURE 5,TYPE OF MAINTENANCE S TRA TEGIES

 Reactive: A run-to-failure approach, when a fault is detected it will be fixed.

o Corrective: Maintenance is a policy to fix and/or replace components either when they have failed or when they are found to be failing. Actions are only performed when a machine breaks down. There are no interventions until a failure has occurred (Bevilacque and Braglia, 2000).

o Detective: (i.e. failure-finding) maintenance applies only to hidden or unrevealed failures and hidden failures usually only affect protective devices (cf. Tsang, 2002). For example: discovering that a fire alarm does not work when you test it.

 Proactive: This approach aims at using models/experience/prognostics to repair before a fault occurs.

o Preventive: overhauling a system before its failure occurs.

(19)

 Predictive: failures are analyzed to discover a possible temporal trend. This makes it possible to predict when the controlled quantity value will reach or exceed the threshold values (Bevilacque and Braglia, 2000).

 Dynamic: Different types of usage of the equipment lead to different amounts of life consumption. Using a dynamic maintenance policy, these variations in usage are linked to deterioration.

o Usage severity based: The variation in usage severity during operation of a system will be monitored. These types of usage are linked to life consumption using physical models (Tinga, 2010).

o Load based: By monitoring the actual relevant internal loads in the component, these loads are linked to life consumption (Tinga, 2010).

 Scheduled: Maintenance conducted on pre-set intervals.

o Usage based: maintenance is scheduled, preventative, maintenance based on usage (production hours, flight hours, etc).

o Time driven: maintenance is scheduled, preventative, maintenance based on a fixed time schedule.

o Opportunistic: this can lead to the whole plant being shut down at set times to perform all relevant maintenance interventions at the same time (Bevilacque and Braglia, 2000).

 Aggressive: The aggressive approach aims at improving the equipment to reduce the number of failures.

o Design improvement: this includes aggressive strategies like for example TPM. The component performance will be improved, resulting in less maintenance needed.

3.2.3 M

AINTENANCE PLANNING

Using the aforementioned maintenance policies, the maintenance should be planned. Maintenance planning is divided into maintenance strategy planning and maintenance task planning. Efficiency of maintenance depends more on the appropriateness of the maintenance strategy planning than on maintenance task planning. Therefore, establishment of a systematic methodology for maintenance strategy planning is an important issue for life cycle maintenance. (Takata et al, 2004). Figure 6 gives an overview of factors influencing the availability of equipment.

FIGURE 6,THEORETICAL AND ACTUA L AVAILABILI TY (BUSSEL &ZAAIER,2001)

(20)

 Availability is the probability that the system is operating satisfactorily. The major difference between reliability and availability is the O&M strategy of the system. A system can be very reliable: i.e. its failure frequency is extremely low, but when no maintenance or repair action is taken after a failure its availability becomes very poor (Bussel and Zaaier, 2001).

 Reliability of a system is the probability that the system will perform its tasks. This probability is usually determined as a percentage of time.

 Maintainability is a more qualitative issue that addresses the ease of repair issue. It can though be expressed in terms of hours needed to complete a repair action.

 Serviceability regards in a similar way the ease of regular (scheduled) service.  Failure is the termination of the ability to perform a required function of a system.

 Accessibility is the percentage of time that a construction can be approached. Evidently the accessibility depends upon the equipment used.

This figure shows that the actual availability depends on the theoretical availability, the accessibility of the site and the maintenance strategy. The accessibility of the site is not very interesting for this study because the CV90 can always be accessed (because it is driven by the crew and they can perform maintenance activities). If this figure is merged with Figure 4, which describes how the maintenance strategy is build up, this results in Figure 7.

Maintenance Strategy Managerial Aspects Technological Aspects Theoretical Availability Maintainability and serviceability Reliability + + Structure and characteristics of facilities Available maintenance technologies Environmental conditions Operational conditions Managerial features of facilities Actual Availability

FIGURE 7,ACTUAL AVAILABILI TY (BASED ON BUSSEL AND ZAAIER (2001) AND TAKATA ET AL (2004))

(21)

varies. This is logic because one would maintain a very unreliable system different than a system which is always operational. Figure 8 shows how the figure changes if this remark is taken into account. Maintenance Performance Maintenance Strategy Managerial Aspects Technological Aspects Theoretical Availability Maintainability and serviceability Reliability + + Structure and characteristics of facilities Available maintenance technologies Environmental conditions Operational conditions Managerial features of facilities

FIGURE 8, MAINTENANCE PERFORMANCE INFLUENCED BY STRATEGY UNDER THE INFLU ENCE OF A VAILABILITY

This figure can therefore be used to gain an understanding about the variables that influence the maintenance performance. It can be concluded from this figure that a change in the maintenance strategy will influence the performance of maintenance. The maintenance strategy has to be selected based on the input of the other variables.

Scheduling

(22)

FIGURE 9,PREDI CTIVE MAINTENANCE (BUTCHER,2000)

3.2.4 M

AINTENANCE

P

LANS

In the past, management in organizations regarded maintenance as “a necessary evil” (Sherwin, 2000). However, in the course of time, maintenance is seen as a value adding activity. Several concepts are developed to design the optimal maintenance function in the company. These concepts can be regarded as being wider than the maintenance strategies because they form approaches for organizational design and sometimes include complete philosophies about how personnel should be motivated. Therefore, these concepts can be seen as a company-wide approach for the design of the maintenance function. As has been said, the relation with maintenance strategies is very close. Several authors see a maintenance concept therefore as a maintenance strategy (see for example: Bevilacqua and Braglia, 2000; Moubray, 1997), other authors regard it as a methodology (e.g. Tsang, 2002). For example Tsang (2002) distinguishes the asset oriented methodology (RCM) from the people oriented methodology (TPM). In this section, two maintenance concepts used in this study will be discussed briefly. This section will only elaborate on the RCM and ECM approach. The ECM approach includes parts of the TPM (total productive maintenance) approach and quality management. Moreover, the selected approaches are mostly focused on the assets whereas for example TPM is a people oriented methodology.

3.2.5 R

ELIABILITY CENTERED MAINTENANCE

RCM originated in the development of a preventive maintenance plan for the Boeing 747 in the 1960s by the maintenance steering group 1 (SMITH, 2004). In a RCM approach, it is not only considered ‘what can be done’, but also ‘why should it be done’. The RCM concept is characterized by four features (Smith, 2004):

1. The primary objective of RCM is to preserve system function

Preserving the output of the system (function) is the primary task of RCM; it does not initially deal with equipment operation. Moreover, it will not be assumed that every item of the equipment is equally important.

2. Identify the failure modes that can cause functional failure

The primary objective is to preserve system function; therefore, specific failure modes of components that can cause functional failures are identified.

(23)

Not all functions are equally important. Therefore not all failure modes are equally important.

4. Select the applicable and effective preventive maintenance tasks

Each prioritized failure mode will be addressed and a preventive maintenance (PM) task considered. The potential maintenance tasks should be judged applicable and effective. Applicable means that the task accomplishes the reason for PM (prevent or mitigate failure, detect onset of a failure, discover a hidden failure). Effective means that the organization is willing to spend the resources to do it. Once a task is found applicable, generally the most effective task is chosen.

3.2.6 E

FFECTIVENESS CENTERE D MAINTENANCE

Effectiveness centered maintenance (ECM) focuses on the service provided to the customer. When a vehicle is broken, the only concern of the user is when the vehicle is working again. This means that the focus of the customer is on the availability of the service instead of the defect rectification time (Pun et al, 2002). In Figure 10, the key elements of the ECM approach are given. The approach focuses on the active participation of personnel, an improvement in the quality of equipment, the development of a maintenance strategy and finally the introduction of a performance measurement system. The contribution of ECM to performance measurement will be discussed in 3.3.

FIGURE 10,KEY ELEMENTS OF ECM APPROACH (PUN,CHIN,CHOW,LAU,2002)

By focusing on the effectiveness of maintenance, ‘doing the right things’ will prevail above ‘doing things right’. The ECM approach includes core concepts of quality management, total productive maintenance (TPM) and RCM.

3.3 M

AINTENANCE PERFORMANCE

To evaluate the process of maintaining a system, a performance measurement system will be created. Performance reflects to what extent the desired results of a particular event are achieved (Vos et al, 2011). Kennerly and Neely (2003) and Lingle and Schiemann (1996) have observed that companies which use a performance measurement system perform better than those who do not measure performance. Moreover, companies can use performance measurement to monitor implementation of new plans and determine if they are successful and how they can be improved (Atkinson, Waterhouse, Wells, 1997). Vos et al (2011) argue that performance measures are needed for four reasons:

1. To learn what current performance is 2. To set overall target levels of performance

3. To evaluate performance and the rewards associated with this performance 4. To manage the performances of processes

(24)

Measuring performance is not only interesting for improvements within the company. For example, outsourcing maintenance to a PBL-Supplier (Performance Based Logistics Supplier) requires measuring the performance the supplier delivers: the maintenance performance. Using a PBL-supplier, the customer will not pay for broken assets (repairs) but pays for a working asset only: pay for usage (Vos et al, 2011). This aligns the goals of the supplier and the customer. The supplier is paid for a certain level of performance (e.g. operational availability, quality, costs) the customer desires. In the other scenario, where the supplier gets paid per repair action, profit is gained when more faults occur. The PBL process requires high (maintenance) supplier involvement. Next to that, the customer (in this case the NL MOD), should be able to assess the maintenance supplier based on the requested level of performance. Therefore, indicators and measurement of performance is needed.

3.3.1 P

ERFORMANCE MEASUREME NT DEVELOPMENT

Parida (2006) argues that the central questions for establishing a maintenance performance measurement system are (1): How should one develop it and what should it look like? And (2): How should one implement and use it?

A performance measurement system should be in line with its corporate and functional strategies and objectives (Bititci, Carrie and McDevitt, 1997). Parida (2006) notes that important issues and challenges associated with a performance measurement system are: relevance, interpretability, timeliness, reliability, validity, cost and time effectiveness, ease of implementation, and updating and maintenance for regular use by stakeholders at various levels. Atkinson et al (1997) list four things a performance measurement system should be able to:

1. Value received: Evaluate if it is receiving the expected contributions from employees and suppliers, the elements of its internal stakeholder group, and the expected returns from the customer group.

2. Value provided: Evaluate whether it gives each stakeholder what it needs to continue to contribute so the company can meet its primary objectives.

3. Process efficiency: Guide the design and implementation that relate to the company’s secondary objectives.

4. Strategic properties: Evaluate planning and contracts.

Vos et al (2011) add a number of requirements to measure performance. They state that performance measurement should be: simple; meaningful; quantifiable, measurable and verifiable; mapped onto a time scale; unambiguous and well defined; achievable; accredited by all parties involved; reflect strategic goals of customer; focus on relevant, critical aspects; attributable to the performance of the organizations involved; should reflect the type of behavior needed (e.g. efficiency, robustness, continuous improvement or responsiveness); the need for leading indicators besides lagging indicators; reflect short term and long term performance.

The latter two requirements, leading and lagging indicators, and short and long term performance are of high importance to the system. Lagging indicators represent past performance, leading indicators provide information which affects performance in the future. Short term performance indicators focus on for example availability. Long term objectives should not be forgotten. A long term indicator can be: system health (or degradation) or durability (Vos et al, 2011).

(25)

by comparing different weapon systems. Therefore, the framework should be suitable to compare different weapon systems.

3.3.2 M

AINTENANCE

P

ERFORMANCE INDICATOR S AND FRAMEWORKS

The performance of maintenance can be regarded on several levels. It can be restricted to the maintenance department only or can include the external effectiveness, which measures the environment influenced by the maintenance department. Moreover, the performance can be limited by focusing on a single system, or measuring the performance of maintaining all systems in a factory. In this section a number of frameworks from relevant literature will be described and their limitations will be discussed.

Swanson (2001) offers maintenance performance measurements based on three criteria: “improvements in product quality, equipment availability and reduction in production costs.” Furthermore, she argues that both proactive and aggressive maintenance strategies have significant positive relationships with these measures of performance. Adamides, Stamboulis and Varelis (2004) assess in their paper the maintenance performance of jet engines in a military context. This is measured by the ratio of: ‘percentage engine availability’ to the ‘total financial and operational costs consumed in achieving this availability at a specific time interval’ (month). This is a good framework to start with, but it does not contain any leading indicators.

Many performance measurement systems are limited to measuring only costs and efficiency (Parida, 2006). The balanced scorecard of Kaplan and Norton (1992) tries to overcome this problem. It focuses on four performance areas: Financial, Customer, Internal processes and Learning & Growth. Parida (2006) proposes a multi-criteria framework which identifies performance indicators in seven critical strategic areas. It claims to be a balanced and integrated framework (as it is partly based on the balanced scorecard) which can be used to achieve total maintenance effectiveness which will contribute to the overall objective of the organization. The framework of Parida (2006) is given in Figure 11.

FIGURE 11,LINKING STRATEGY, TOTAL MAINTENANCE EF FECTIVENESS AND MAINTENANCE MEASU RING CRI TERIA (PARIDA,2006).

(26)

Next to the given performance indicators, Table 3 gives an overview of performance measures. This overview was created in a literature review of maintenance performance measurement by Simões, Gomes and Jasin (2011).

Equipment losses Breakdowns Cycle time (delivery) Efficiency

Events Flexibility MTTF Inventory cost

Manpower Service level Time Tools

Workorders Human resources Breakdown Maintenance Labor costs

Defects Downtime costs Maintenance organization Preventive maintenance

Accidents Maintenance strategies Spare parts Productivity

Reliability Failures Costs Downtime

Equipment Materials MTTR Tasks

MTBF Quality Availability OEE

TABLE 3,TYPE OF MAINTENANCE PERFORMANCE MEASU RES (SIMÕES,GOMES,JASIN,2011)

Moubray (1997) argues that maintenance can be viewed from two viewpoints. Firstly, focusing on how well maintenance ensures that users can use their assets in the way they want to do. This first viewpoint is referred to as: maintenance effectiveness. The second viewpoint concentrates on how well the resources for maintenance are used. This is referred to as: maintenance efficiency. These two factors together form the maintenance performance.

EFFECTIVENESS AND EFFICIENCY

To evaluate the contribution of maintenance to the performance of equipment, the effectiveness of each function has to be measured on an on-going basis. Therefore, a bright understanding of each function and understanding of what is meant when it is said to be ‘failed’ is required. The arbiter of effectiveness is the user, who should have realistic expectations (Moubray, 1997). One should note that equipment performance is different from equipment effectiveness because equipment performance measures functional effectiveness. This distinction means that the focus is on ‘what the equipment does’ instead of ‘what the equipment is’ (Moubray, 1997).

The efficiency side of the maintenance performance consists of: the maintenance costs, labor, spares and materials, and planning and control (Moubray, 1997).

Maintenance effectiveness consists of the variables: how often it fails (reliability), how long it lasts (lifespan), how long it is out of service when it does fail (downtime), how likely it is to fail in the next period (dependability), and the efficiency measured as the ratio of how well something is performing relatively to how well it should be performing (Moubray, 1997). Parida (2006) argues that a maintenance performance measurement system should measure both the internal as well as the external effectiveness (Figure 12). However, these criteria should be tailored on the required focus of the measurement and the environment of the maintenance function.

(27)

FIGURE 12,TOTAL MAINTENANCE EFF ECTIVENESS (SOURCE:PARI DA,2006)

COS TS OF MAINTENANCE

To assess the costs which are associated with performing maintenance, the total costs of maintenance should be made visible. Conforming to the concept of ‘costs of poor maintenance’ (CoPM), maintenance costs can be divided into four categories. This concept provides a viewpoint for the identification of deficiencies in the maintenance performance. In this concept, a distinction has been made between costs of conformance and costs of non-conformance. All costs that contribute to expected deliveries are seen as the costs of conformance.

FIGURE 13,COSTS OF POOR MAINTENANCE (COPM) CONCEPT (SALONEN AND DELERYD,2011)

As can be seen in Figure 13, costs are created when preventive maintenance is not, poorly or unnecessary performed. Moreover, poor equipment reliability leads to unnecessary costs. Therefore, it is very important to keep track of the costs of maintenance. Furthermore, Salonen and Deleryd (2011) have noted that the costs of non-conformance maintenance activities are the highest of the two to firms (see Figure 14). They argue that a company should use this concept to identify the weaknesses of their maintenance performance and try to minimize the costs of non-conformance.

(28)

3.3.3 L

IMITATIONS OF EXISTI NG FRAMEWORKS

Now a number of frameworks for the measurement of maintenance performance have been discussed, first the limitations and benefits of these frameworks will be discussed.

Firstly, the discussed frameworks are mostly tailored to private-sector companies. For example: Figure 12 of Parida (2006) includes measurements like ‘growth in market share’. This type of parameters are not applicable in a military context since using a weapon system does not lead to profit from sales or a market share. One of the most characterizing elements of the military context is that the value of higher availability is hard to measure, especially in a peace situation. The value attached to availability could also differ. In wartime, availability might be very important, but in peacetime, costs might be of higher interest to a defense department. Performance measurement is quickly limited to the internal effectiveness (costs). However, also in a military context, the effectiveness of the maintenance process should be added to the performance measurement system to guarantee enough operational available systems. Secondly, a number of frameworks, for example: Swanson (2001) and Adamides et al (2004) only include lagging indicators. These frameworks can be used to assess performance of the past, but have no value for the prediction of performance in the future. Thirdly, the framework of Parida (2006) offers a total overview on the maintenance process. However, as Artley and Stroh (2001) argue, one of the major pitfalls of a measurement system is measuring to many parameters. This will result in ignorance of the managers and employees. Moreover, it becomes a day-task to measure all these parameters. Finally, to be able to use the frameworks for benchmarking purposes, as is recommended by Simoes et al (2011), the parameters shouldn’t be single values but ratios.

Therefore, the correct parameters for measuring the maintenance performance should be selected. The most applicable parameters will be selected to form a framework for measuring the maintenance performance of a military system. Selection criteria for the proposal of a framework capable of measuring maintenance performance in a military context will be given in section 3.5.

3.4 DYNAMIC MAINTENANCE

Wubben (2009) argues that by enhancing a dynamic maintenance strategy the maintenance will be tailored better to the actual usage. This could result in a better cost-effectiveness of the system. This is partly comparable to an improvement in the maintenance performance. Therefore, the type and severity of usage should be related to the maintenance need of equipment. This type and severity should be combined in so called usage profiles. Such a profile should be based on several parameters which represent the causes of the wear and tear of the equipment. An operator of the equipment could, after using the vehicle, note how long the equipment is used in each usage profile. Next to that, information given by the operator could be enriched by information retrieved from a possible vehicle information system which registers certain usage parameters. This section will discuss already existing concepts in the field of usage and health monitoring.

Usage monitoring involves automated tracking of life-limited parts and retirement of these parts based on actual usage rather than worst-case conservative usage estimations (Romero, Summers, Cronkhite, 1996). This means that if a part is used less severe, actual life-time could increase. This idea is visualized in Figure 15.

(29)

and operational environments; and previous component and system level testing and maintenance (Farrar and Lieven, 2006). Furthermore, Farrar and Lieven (2006) argue that it is important to distinguish between health and usage monitoring. They define these concepts as:

 Health monitoring: The process of identifying and quantifying the extent of damage in a system based on information extracted from the measured system response.

 Usage monitoring: The process of acquiring operational loading data from a structure or system, which preferably includes a measure of environmental conditions (e.g. temperature and moisture) and operational variables such as mass or speed.

FIGURE 15,POTENTIAL BENEFITS PROVIDED BY USAGE MONI TORING WI TH HU MS (ROMERO,SUMMERS AND CRONKHI TE,1996) A system often used in rotorcraft is HUMS (Health and Usage Monitoring System). Heine and Barker (2007) define HUMS as: “a system of sensors, processors and algorithms that give an indication of remaining component life”. When HUMS would be applied to military vehicles, these sensors will be exposed to rough terrain, extreme temperature fluctuations, dust and moisture, and need therefore to be very reliable (Heine and Barker, 2007). Heine and Barker (2007) argue that available sensors for use in aircraft, plants, or electronic applications would not survive long in a military vehicle.

Farrar and Lieven (2006) argue that both health and usage monitoring are required for proper damage monitoring. This process starts with measuring the current condition of the system (health) and consequently monitoring the usage of the system. Multiple interpretations could be given to a health and usage monitoring approach. Figure 16 shows a hierarchy of different prognostic interpretations. From this figure can be deduced that statistical methods are most simple and have the lowest costs. However, using a physical model will give the highest accuracy.

FIGURE 16,HIERARCHY OF PROGNOS TIC APPROACHES (LEBOLD AND THURSTON,2001)

(30)

Monitoring the usage of the system, coupled to a thermal / fluid / structural model leads to an internal load. This load will be coupled to a failure model. The result of this calculation is knowledge about the (predicted) condition of the system.

FIGURE 17,SCHEMATIC REPRESENTA TION OF THE RELA TION BETWEEN USAGE, LOADS, CONDITION AND LIFE CONSU MPTION.THE MOST IMPO RTANT RELATIONS ARE (1) THE USAGE-TO-LOAD AND (2) THE LOAD-TO-LIFE RELATIONS (TINGA,2010)

Figure 18 shows the basis for a usage load profile. A usage load profile is primarily based on the operational context and the mission. The operational context describes a number of environmental conditions like temperature and humidity.

(31)

3.4.1 M

ETHOD SELECTION

When applying a dynamic maintenance strategy, one should be able to have the right focus. Therefore, reliability centered maintenance (RCM) and a platform degrader analysis, will be used to ensure a methodological well-founded approach of implementing a dynamic maintenance strategy. Firstly, by using the RCM approach of Moubray (1997). This approach consists of seven central questions:

I. What are the functions and associated performance standards of the asset in its present operating context?

II. In what ways does it fail to fulfill its functions? III. What causes each functional failure?

IV. What happens when each failure occurs? V. In what way does each failure matter?

VI. What can be done to predict or prevent each failure?

VII. What should be done if a suitable proactive task cannot be found?

In addition to the RCM approach the platform degrader analysis of Banks, Reichard, Hines and Brought (2008) can be used. This platform degrader analysis is based on the RCM approach of Moubray (1997). The degrader analysis aims to determine which platform components and subsystems contribute the most toward the loss of vehicle operational availability and then identify diagnostic, predictive and prognostic technologies that are mature and appropriate to apply to these specific components and sub-systems (Banks et al, 2008). A major advantage of this approach is that it focuses on the top candidates for health monitoring, rather than conducting full FMECAs on each platform (Banks et al., 2008). This ensures the in-depth focus of any study, instead of providing superficial recommendations for many components. The platform degrader analysis consists of three steps:

I. Identify components which have the lowest reliability and greatest number of maintainability issues

II. Evaluate how these components fail and determine their dominant and critical failure modes (using FMECA on only the top degrader components and subsystems)

III. Identify appropriate technology solutions for monitoring each dominant and critical failure mode, capable of providing an on-board diagnostic, predictive or prognostic assessment.

TECHNICAL VERSUS BUS I NESS APPROACH

ASTER T HESIS M

M

ASTER

T

HESIS

Public Version

T

OWARDS IMPROVING MAINTENANCE PERFORMANCE

A

BSTRACT

P

REFACE

Table of Contents

G

LOSSARY OF TERMS

1. I

NTRODUCTION

1.1 R

1.2 R

1.3 S

1.4 S

1.5 L

-QUESTIONS AND CHAPTERS

1.6 T

2. R

ESEARCH

M

ETHODOLOGY

2.1 R

2.2 C

CV90 CASE S TUDY

2.3 C

3. L

ITERATURE REVIEW

3.1 M

3.2 M

3.2.1 D

3.2.2 M

3.2.3 M

3.2.4 M

P

3.2.5 R

3.2.6 E

3.3 M

3.3.1 P

3.3.2 M

P

3.3.3 L

3.4 DYNAMIC MAINTENANCE

3.4.1 M