Maintenance strategy development

(1)

Maintenance strategy development

Development and application of a maintenance policy decision-making framework at Scania Production Meppel B.V.

University of Groningen Faculty of Economics and Business MSc. Technology and Operations Management

23-01-2015

Author: Supervisor:

P.C. van Eck Dr. Ir. W. Alsem

(2)

Abstract

(3)

Preface

This Master thesis is the final stage of my Master degree Technology and Operations Management at the University of Groningen. The past six months I have conducted research regarding the development of a maintenance strategy in manufacturing facilities in cooperation with Scania Production Meppel and the University of Groningen. While writing this thesis, I had the opportunity to combine scientific and practical knowledge into a combined methodology and apply it aiming to improve the current maintenance strategy.

I would like to thank all participants contributing to the realisation of this thesis. In particular I would like to thank Stefan Smit, Sjoerd Kunnen, Henk Loijenga of Scania Meppel and Wilfred Alsem of the University of Groningen for the supervision and support in successfully completing this thesis. Also a special thanks to the maintenance department of Scania Meppel for their efforts.

(6)

2

1. Introduction

Checking tire pressure, changing a worn-out tire, and replacing the spark plug are all maintenance activities in order to keep your motorcycle on the road. Without these activities, your motorcycle might wear out faster or break down altogether. This also applies to Scania Production Meppel (Scania Meppel), where complexity,costs and consequences are considerably higher. Therefore, in order to control a facility’s performance and costs, a maintenance strategy should be developed and executed. However, whether Scania Meppel currently attains a maintenance strategy which is both effective and efficient, is unclear. The importance of a maintenance strategy is emphasised by the fact that maintenance costs can reach up to 70 percent of the total production costs (Belvilacqua and Braglia, 2000; Wang et al., 2007), of which up to one third are wasted resulting from unnecessary or improper activities (Mobley, 2002; Wang et al., 2007). From this, traditional views on maintenance encompassed maintenance as a necessary activity, costing money and time. Still, Al-Najjar and Alsyouf state that maintenance is a profit generating function (2002; 2004; Alsyouf, 2007).

According to Pintelon et al. (2000) maintenance is “the set of activities required to keep physical assets in the desired operating condition or restore them to this condition.” In order to fulfil this function, maintenance objectives are to maximise reliability and maintainability, influencing both availability and quality (Hopp and Spearman, 2008 and Kothamasu et al., 2006). Depending on its efficiency and effectiveness in attempting to optimise this availability, maintenance can become both a cost and profit centre. From this, costs are added to the objective function of maintenance.

In order to bring about a desired situation, thus for maintenance to ultimately increase availability, a maintenance strategy is developed. According to Kelly (1997), a maintenance strategy entails the identification, researching and execution of repair, replacement and inspection decisions. Alsyouf (2007) states that a maintenance strategy entails the policy (‘best’ life plan) for each unit of the manufacturing facility, describing what events should prompt which type of maintenance action at what time.

For framing these maintenance policies into a maintenance strategy, different decision frameworks have been drawn up, as the Balanced Score Card by Kaplan and Norton (1992) and the Analytical Hierarchy Process (Saaty, 1980). However, current frameworks are also criticised for e.g. simplicity, inefficiency and lack of practical application. Also, Garg et al. (2006) states that limited work has been directed towards developing an operational decision support system using an appropriate maintenance optimisation model.

(7)

3 The objective of this Master Thesis is to identify improvements to the current maintenance strategy, aimed at improving the availability-cost trade-off of Scania Meppel’s production process. For the development of an improved maintenance strategy, different decision-steps need to be executed. However, as previously stated, current maintenance policy decision-support frameworks (or systems) are limited. Therefore, a sidestep needs to be made, developing an appropriate and holistic maintenance policy decision-making framework. Finally, using this framework, recommendations are developed towards improving the current maintenance strategy, answering the following research question:

“Which maintenance strategy contributes most to the availability-cost trade-off of Scania Production Meppel’s production process?”

It should be noted that a maintenance strategy is viewed as a holistic set of maintenance policies for different processes in order to optimise the availability of the production process. In the table below, a reading guide is shown.

Reading guide

Maintenance strategy development and application Complete proposal

Organisational setting Chapter 2, paragraph 2.1, p.4, appendix A, p. 45, appendix, B, p. 46, appendix K, p. 60-63

Theoretical setting Chapter 2, paragraph 2.2, p 5-8, appendix C, p. 47-48 Research question and design Chapter 3, p. 9-10, Chapter 4, p. 11-13

Maintenance policy framework development Chapter 5, p. 14-17

Case study Scania Meppel Chapter 6, p. 18-38

Maintenance policy decision-making Chapter 6, p. 34-38

Maintenance policy recommendations Chapter 6, p. 36-38, appendix M. p. 67-68

Conclusion Chapter 7, p. 39

Discussion Chapter 8, p. 40

References p. 42-44

Appendices p. 45-68

(8)

4

2. Research framework

The research framework describes the underlying setting of this study, both from an organisational and a theoretical perspective. Paragraph 2.1, covers the organisational background, describing in which setting this study is executed. Next, in paragraph 2.2, the theoretical background, the underlying theory and purpose of maintenance and its objective is discussed. Furthermore, the maintenance strategy and maintenance policies are covered. In paragraph 2.3, maintenance policy development methods are discussed on their pros and cons, executing an initial exploration.

2.1 Organisational background – Scania Production Meppel

At Scania Meppel, cabin parts of trucks (appendix A) are varnished for Scania Productions in Zwolle in The Netherlands, Södertälje in Sweden, Angers in France, and third parties. Scania incorporates many Lean principles and methods and uses many manufacturing techniques of JIT. Scania incorporated these philosophies, techniques and methods into the Scania Production System (SPS). Scania Meppel strictly works with a make-to-order production system and strives to satisfy customer’s (demand), respect its employee and eliminate waste, ultimately in order to ensure continuity and growth. Furthermore, Scania Meppel has a product-layout, with production lines T1, T2 and T3 distinguished through e.g. primer or water-based coatings. For a general process overview, refer to appendix B. The production facility works in 2 production and 3 maintenance shifts, leaving the third shift for more substantial maintenance activities possibly requiring downtime. Regarding performance, in 2015, Scania Meppel aims for a minimal OEE (Overall Equipment Effectiveness; appendix B) of the production process of 82.5%, and 90% or higher for 2018. Today’s OEE average of Scania Meppel is approximately 78%, indicating improvements are necessary in order to reach these goals. The OEE is built up from quality, availability and performance efficiency.

In the current situation, maintenance activities at Scania Meppel are performed on a reactive, preventive, and predictive way. Many of these features are intertwined with TPM (Total Productive Maintenance) principles as autonomous maintenance, movement towards proactive maintenance, and a focus on improving the OEE. During the morning and day shifts, maintenance activities focus primarily on reactive maintenance, where occurring problems are handled ad hoc, focused on keeping the production facility in working order. During the third shift, maintenance is comprised of maintenance activities resulting from events which have occurred during the day and activities which originate from proactive maintenance (time- or measurement-based). These judgements whether to act reactive or proactive has been developed over time, primarily based on experience of specialists and recommendations of the supplier of the facility. However, currently no standardised method is in place for decision-support regarding the development and evaluation of maintenance policies.

(9)

5 2.2 Theoretical background

According to Hopp and Spearman (2008), a ‘good’ return on investment (1; ROI, see for relations figure 1) in the long run is regarded as the fundamental objective of a manufacturing facility since it satisfies the various stakeholders, employees and customers. In production environments, variability negatively influences the throughput, assets and costs (2), hence, a company’s ROI and should therefore be kept low (2008). In order to absorb variability, and therefore disruptions, inventory, capacity and time are used as buffers (3; Hopp and Spearman, 2008). Through either resolving or maintaining assets, maintenance plays a significant role in minimising disruptions.

2.2.1 Maintenance and maintenance objectives

Maintenance entails, according to Pintelon et al. (2000), “the set of activities required to keep physical assets in the desired operating condition or restore them to this condition.” This is confirmed by Wiegand et al. (2005) who state that maintenance allows value to be added (through availability) and that maintenance should strive for process stability and quality, preventing waste in the process. This since maintenance attempts to ensure availability and quality (4) (Hopp and Spearman, 2008 and Ben-Daya and Duffuaa, 2010), enabling the system be to create the desired throughput, decrease assets (e.g. inventory) and costs (e.g. scrap and overtime) (2). In this perspective, maintenance costs can be perceived both directly and indirectly. Direct maintenance costs regard e.g. personnel and material costs (6), where indirect costs originate from manufacturing performance (availability and quality of finished products). Direct maintenance costs presents an increasing limiting factor, thus is subordinate to availability, on the maintenance process in the pursuit of improving availability, including quality of equipment, and indirectly quality of products.

In maintenance, availability is the actual time available for production, hence the total available time minus the unplanned outages. Quality can be viewed both from the perspective of the product (rework) and quality of the maintenance activities. Through increased automation, the importance of equipment maintenance in controlling quality is of increased importance (Ben-Daya and Duffuaa, 2010). Quality of maintenance activities affect equipment performance and consequently the quality of the final product (Ben-Daya and Duffuaa, 2010). Therefore, quality of equipment and quality of the final product can be distinguished. For example, if a spray-paint robot is insufficiently maintained and therefore falter or fails periodically, the process will experience speed losses and/or a lack of precision, resulting in product. Kothamasu et al. (2006) confirms this, stating that quality is a maintenance objective in order to maximise availability (quality of equipment) and, although indirectly, minimise defective output (quality of finished products) to increase the throughput, preferably with less assets and costs. Therefore, quality of equipment is regarded as an integral part of reliability and consequently availability.

(10)

6 attempts to ensure high quality of equipment, consequently improving the availability and indirectly the quality of finished products (4).Due to the unquantifiable nature of the influence of maintenance to the quality of products, only quality of equipment is included in this study, since this is the only element of quality which is directly influenced by maintenance.

2.2.2 Maintenance strategy

In order to reach or strive towards the maintenance objectives stated above, a maintenance strategy is developed. As described in the introduction, a maintenance strategy entails the identification, researching and execution of repair, replacement and inspection decisions (Kelly, 1997). Alsyouf (2007) continues, arguing that a maintenance strategy involves formulating the ‘best’ life plan (or maintenance policy) for each unit of the plant concerned, and also to formulate an optimal maintenance schedule for this plant, describing what events should trigger what maintenance action. Muchiri et al. (2010) and Tsang et al. (1999) add that for maintenance to truly be effective, a holistic approach is advocated. Therefore, the strategic maintenance objectives should be aligned with the strategic objectives of production as well as the objectives of the corporation as a whole (i.e. Tsang, 1999, Muchiri et al., 2010 and Horenbeek and Pintelon, 2013). An effective maintenance strategy manifests itself by a combination of effects in throughput, assets and costs, increasing the difference between input and output, consequently increasing return on investment.

In addition, Márquez et al. (2009) state that whatever strategy is selected, it should evolve, or as Waeyenbergh and Pintelon (2002) described it, to be periodically reviewed, in order to retain its usefulness against changes in the manufacturing facility. Moreover, Wang et al (2007) state that an optimal maintenance strategy is necessary in order to increase reliability and maintainability of a manufacturing facility without the necessity of large investments.

Again, it should be noted that when developing a strategy, more than one maintenance policy can be attained at the same time. From this, Alsyouf (2007) concludes that it is highly important to know which maintenance policy is the most cost effective and suits the given system in its operating context. In figure 1, depicts that a maintenance strategy, composed of different maintenance policies, (should) positively influence reliability and maintainability, consequently availability and indirectly the quality of finished products, which, in turn, positively influence the throughput, assets and costs and ultimately the return on investment. Indirectly, an increased availability and quality of products will positively influence the buffers within a manufacturing facility, decreasing them. This relation supports the statement of Alsyouf (2007) that maintenance is a profit centre.

(11)

7 2.2.3 Maintenance policies

In literature, maintenance policies are also referred to as approaches, programs, concepts, techniques or strategies. Maintenance policies can be divided in three categories, namely corrective, preventive (Kothamasu et al., 2006) and aggressive maintenance, focussing on actually improving the function and design of the production equipment (Swanson, 2001). First, reactive maintenance entails executing inspection, repair, or replacement actions after failure (Kelly, 1997). Proactive maintenance are maintenance activities which are carried out before failures occur, aiming to prevent failures, consequently elongating availability. Proactive maintenance can be subdivided in preventive and predictive maintenance, which act on respectively time-based intervals and actual condition and performance. Next, aggressive maintenance concepts, also referred to as process oriented ‘holistic’ approaches are comprehensive approaches as for example RCM and TPM. Opportunistic maintenance (OM) is added, which is a policy for a collection of components aiming to increase efficiency of the maintenance function (Koochaki et al., 2011; Nicolai and Dekker, 2008). Next, redundancy is for example machines who act as a double (backup) in case failures occur. Finally, Design Out maintenance is a once-off action, designing or redesigning a given process in order to prevent the need for maintenance altogether. In the figure 2 an overview of maintenance concepts are categorised according to the nature of maintenance based on Koochaki et al. (2011), Waeyenbergh and Pintelon (2010), Alsyouf (2007), Kothamasu et al. (2006), Wiegand et al. (2005), Swanson (2001), and Kelly (1997). In appendix C, the different maintenance policies mentioned above are elaborated.

Figure 2 - Maintenance concepts

2.3 Maintenance policy development methods

In order to develop an effective maintenance strategy, each system within the organisational setting needs to be identified. For these systems, a maintenance policy is developed. From literature, many different methods have been developed in order to find an appropriate maintenance policy. Therefore, in this paragraph, the different maintenance policy development methods are discussed, identifying advantages and shortcomings of methods in an attempt to find a combination of methods utilising advantages and overcoming shortcomings.

(12)

8 1. Description of a technical system, its function and criticality within the manufacturing facility; 2. A modelling of the deterioration of the system in time and possible consequences for this

system;

3. Identification of available information regarding the system and actions open to management;

4. The use of an objective function and an optimisation technique, striving for the most optimal trade-off.

A well-known method, developed by Saaty (1980), is the Analytic Hierarchy Process (AHP), which uses different weights coupled to the different selection criteria’s in order to select the ‘best’ maintenance policy. Furthermore, Cassady et al. (2001) developed a mathematical programming framework in order to support decision making focused on determining the optimal subset of maintenance activities. Next to the AHP, different multi criteria decision making (MCDM) methods have been proposed by, among others, Al-Najjar and Alsyouf (2002) applying a fuzzy MCDM and Ahmadi et al. (2010) proposing the combination of the AHP, TOPSIS, and VIKOR methods. Also different additions and variations to the AHP have been developed by for example Horenbeek and Pintelon (2013) proposing the ANP (Analytic Network Process), considering all operational levels (strategic, tactical and operational level). Also, Wang et al. (2007) propose adding fuzzy linguistics to the AHP, adding the inclusion of qualitative criteria within the method. In addition, different frameworks were developed, proposing different decision-steps, in order to systematically, incorporating different methods as the AHP, reach a maintenance policy selection. Márquez et al. (2009) developed a framework supporting the formulation of a maintenance strategy for complex manufacturing plants.

The listing above is merely a selection from different developed, often similar, methodologies and frameworks. From literature, these methodologies and frameworks are also widely criticised by, for example, Ishizaka and Nemery (2014) who state that MCDM is a time-consuming, inefficient and often unfeasible methodology. Other remarks are, for example Muchiri et al. (2010), who states that for maintenance to be effective, the objectives of maintenance need to be aligned with both corporate and manufacturing objectives. This is supported by Tsang et al. (1999), promoting a holistic approach to maintenance policy selection. Furthermore, Muchiri et al. (2010) state that performance indicators of the maintenance process should be attained in order to be able to control and steer maintenance according to its objectives. Finally, Márquez et al. (2009) state that whatever model is developed or selected, it should continuously evolve in order to stay useful against the fast changes that occur in business, communications and industry.

(13)

9 From the literature review discussed above, it has become clear that different approaches each have different advantages and shortcomings. In order to surpass these shortcomings, a framework needs to be developed, selecting appropriate decision-steps and corresponding methods within this framework. Furthermore, additions to existing frameworks possibly need to be made in order to surpass the shortcomings mentioned in literature. Therefore, the framework should be holistically focused, taking consequences of different approaches into account. Furthermore, two input streams are identified. These streams are the maintenance objectives and the equipment performance of the maintenance process (Muchiri et al., 2010). The maintenance objectives need to be aligned to both the corporate and manufacturing objectives and prioritised according the strategic, tactical, and operational level (aggregation levels). The second information stream consists of maintenance measures obtained from equipment performance, focused on the availability-cost trade-off. Finally, the framework will need to be validated and improved continuously, preserving the utility and effectiveness of the framework and ultimately the maintenance strategy.

3. Research question

From the research framework, the variables and their relations, depicted in the figure below, were identified. As stated, maintenance objectives (1; see figure 3) are to be aligned to the corporate and manufacturing objectives (2). Furthermore, maintenance measures from the production process (equipment performance; 3) can be used in order to identify performance gaps, possibly identifying maintenance improvement possibilities (Muchiri et al., 2010). Both the maintenance objectives and the equipment performance form input in the maintenance policy development (4). The policy development is a composition of decision-steps which are to be identified and embedded in a maintenance policy decision-making framework. Such standardised framework is not yet developed in literature or in practice. The collection of these policies resulting from this framework, form the maintenance strategy (5). A maintenance strategy consequently influences the allocation of resources, influencing reliability, maintainability (6) and the direct maintenance costs (8).

(14)

10 Figure 3 –System of interest; maintenance strategy effects and inputs

From the figure 3, it has become apparent that when maintenance policies are and consequently a maintenance strategy is developed, the trade-off between availability and direct maintenance costs is central. Furthermore, the decision steps which together form a comprehensive maintenance policy decision-making framework need to be identified and combined.Subsequently, the goal of this study is to develop an improved maintenance strategy, aimed at improving the availability-cost trade-off of Scania Meppel’s production process. Based on this goal, the following research question has been composed:

“Which maintenance strategy contributes most to the availability-cost trade-off of Scania Production Meppel’s production process?”

As stated before, in order to find improvements for the current maintenance strategy, a maintenance policy decision-making framework should be developed. This way, the current situation of the maintenance strategy of Scania Meppel is identified, followed by the desired state of affairs and consequently the discrepancy between these two states.

From this and from the preceding chapters, sub-questions have been developed in order to systematically reach a satisfying answer in both practical and theoretical sense. For this, the sub-questions below have been divided in sub-sub-questions regarding the development of the maintenance policy decision-making framework and the development of a maintenance strategy at Scania Production Meppel.

Sub-questions maintenance policy decision-making framework:

1. Which decision steps are necessary in order to form a comprehensive maintenance policy decision-making framework?

2. Which methods should be incorporated in order to effectively execute this framework? Sub-question development of a maintenance strategy at Scania Meppel:

3. What is the current maintenance strategy of Scania Meppel?

4. What is the current equipment performance (availability-cost trade-off) of the production process of Scania Meppel?

5. What maintenance policies are preferable under which circumstances for Scania Meppel? 6. Which maintenance policies would, and to what extent, compared to the current policies,

(15)

11

4. Research design

The aim of this six month study is to develop and optimise a maintenance strategy, composed of different maintenance policies. This objective is aimed to be accomplished by the development and application of the maintenance policy decision-making framework. Three main steps are to be executed in order to ultimately develop a maintenance strategy, which are framework development, framework application at Scania Meppel, and strategy development.

4.1 Framework development

Currently, a comprehensive decision-making framework regarding maintenance strategy development is not predefined not at Scania, nor in literature. Therefore, a framework combining different rules and methods, as discussed in paragraph 2.3, containing all decisions needed is developed. Based on the sub-questions 1 and 2, stated in chapter 3, the following steps in the development of the framework are to be executed:

1. Identify maintenance policy decision-making steps; 2. Identify methods per step;

3. Identify data input needed per decision-making step.

In order to create a framework for identifying the maintenance policy decision-making steps and appropriate supporting methods, theory building is used. Using this approach, constructs, methods and mechanisms from literature are reviewed, while taking the available data into account. Using theory building, the decision steps are identified, including appropriate methodologies and mutual linkages, consequently answering sub-question 1 and 2. Input for theory building is both a literature study and semi-structured interviews.

4.2 Framework application

The second part of this study is the actual application of the developed framework at Scania Meppel. The output from the implementation of these decision-steps is twofold. First, maintenance policies will be selected next to a verification effort of the framework. The following steps will be executed when applying the framework at Scania Meppel:

1. Identify the current maintenance policies (strategy);

2. Measure the current equipment performance of the production process; 3. Identify preferences regarding maintenance policy selection;

4. Select the most beneficial maintenance policies.

In order to execute these steps, theory testing is used at Scania Meppel using a case study. For this, both qualitative and quantitative research approaches are used, taking both e.g. statistical production performance data and expert’s knowledge into account, validating the use of a case study (Eisenhardt, 1989; Karlsson, 2009). Furthermore, using this case study, the framework is applied and validated on expected and the actual behaviour of the framework.

(16)

12 experts from the maintenance department. Next, the alternative policies are reviewed from the perspective of Scania Meppel, identifying a classification of preferred policies (sub-question 6). Using this classification, each policy can in turn be considered. Finally, the most appropriate policy is chosen in accordance to technical and economical restrictions, directed towards improving the availability-cost trade-off and consequently the maintenance strategy, answering sub-question 7. Methods in order to answer these sub-questions are semi-structured interviews, data analysis (ERP), and literature research. Furthermore, production and maintenance is observed in order to attain a deeper understanding of the mechanisms, reasons and choices regarding the maintenance activities.

4.3 Strategy development

Finally, combining each developed maintenance policy, an improved maintenance strategy is attained. For this, the current maintenance strategy and its performance is identified and validated. Next, after developing the maintenance policy decision-making framework, improvements to current maintenance policies are identified and recommended in order to further improve the current availability-cost trade-off. However, due to time limitations, only recommendations regarding the maintenance strategy are made within this study. The actual application and evaluation (through equipment performance measurement) of this newly developed maintenance strategy is regarded out of scope.

Throughout this study, validation is of great importance, ensuring correctness of the decision-steps and its methods throughout the study. This validation consists of feedback to both framework development and framework application, checking the correctness of assumptions used in the framework. For example, from Karlsson (2009), does the applied performance measure what they intent to measure (construct validity)? Can findings be generalised (external validity)? And, are causal relationships valid (internal validity)?

4.4 Deductive research design

(17)

13

Triangulation

Case study:

- Observations

- Semi-structured interviews - Historic data analysis

Theory building: - Literature study - Semi-structured interviews Strategy development (3. result) Framework development (1. rule) Framework application (2. observation) Decision steps Validation Validation input input Policy mix Validation Out of scope Figure 4 - Deductive research design

4.5 Scope

Both from this chapter as in chapter 3, a number of limitations have been stated. First, indirect maintenance costs are not included since the allocation of costs to the maintenance function originating from the production process is subject to interpretation. Next, quality and performance, part of the OEE, are regarded out of scope due to their unquantifiable nature and difficulty to pinpoint the exact influence maintenance has on these indicators. In addition, the strategy development, shown in figure 4, is out of scope. Furthermore, due to time limitations, the validation (evaluation) of the actual developed strategy originating from this study is regarded out of scope. The reason is that in measuring the actual performance of a newly implemented strategy it takes time to reach a reliable data set. Finally, the strategy development is somewhat limited in this study. Although the actual strategy developed is included within this study, the planning, executing, evaluating and improvement (Plan, Do, Check, Act) of the developed strategy is regarded out of scope. In addition, the maintenance strategy should constantly be subject to evaluation in order to ensure continuous improvement towards optimising the availability-cost trade-off.

4.6 Theoretical and practical contributions

A theoretical and practical contribution to existing literature is made by using both theory building, developing a decision-making framework, and a case study at Scania Meppel, where current literature regarding maintenance strategy selection, its criticism and remarks are taken into account.

(18)

14

5. Maintenance policy decision-making framework

A maintenance policy making framework entails a structure which supports the decision-making of a maintenance policy of a given system and its components. From the literature discussion in paragraph 2.3, maintenance policy development methods, it has become apparent that numerous decision-making methodologies exist for developing a maintenance strategy. However, these methodologies are also widely criticised for e.g. simplified and delimited area of application. In order to create a standardised framework applicable for the selection of an appropriate maintenance policy at Scania Meppel, different making steps are identified and combined. Such decision-steps were developed by Waeyenbergh and Pintelon (2002), who proposed a general maintenance policy decision-making framework for the application to complex technical systems. This framework was originally developed for developing a maintenance strategy for paint-spraying robots at a truck cabin paint shop. Due to its general nature, the framework of Waeyenbergh and Pintelon (2002) is limited but also highly customisable. Furthermore, the decision-steps incorporate both qualitative and quantitative knowledge when analysing a maintenance strategy. This framework is therefore used as a basis, complemented by input from other literature in order to cover current limitations. In order to create a maintenance policy decision-making framework which can be effectively and efficiently be applied at Scania Meppel, different customisations are needed. First, as suggested by Muchiri et al. (2010), maintenance objectives need to be identified, based on both the corporate and manufacturing objectives. This integration is done in order to further elaborate performance measured compared to the policy selected. Furthermore, from Koochaki et al. (2011), opportunistic maintenance is considered when optimising the selected maintenance policies. Finally, from Márquez et al. (2009), continuous improvement is added, in order to ensure the usefulness of the strategy in the current situation. Furthermore, a limitation to only focus on insourced activities has been made in order to focus solely on the core business of the maintenance function of Scania Meppel. In figure 5, the maintenance policy decision-making framework is schematically depicted. In short, this framework is based on the 7-step CIP framework developed by Waeyenbergh and Pintelon (2002), Wang et al. (2007), Márquez et al. (2009), Muchiri et al. (2010), Koochaki et al. (2011) and input from Scania Meppel. Including the customisations mentioned above, the maintenance policy decision-making framework consists of the following basic steps:

1. Identify corporate and manufacturing objectives (1, 2; see figure 5);

2. Identify maintenance objectives, resources and performance targets (3, 4, 5, 6); 3. Identify Most Important Systems (7; FAHP);

4. Identify Most Critical Components (8; FMECA);

5. Maintenance policy decision per component (9; decision tree);

(19)

15 Figure 5 - Maintenance policy decision-making framework

The first step in this framework is to identify the objectives of the maintenance function (3; figure 5), aligned to both corporate (1) and manufacturing objectives (performance requirements, (2)). Next, the available resources of the maintenance function are identified (4) in order to align (5) the objectives to the targets of maintenance (6) (Waeyenbergh and Pintelon, 2002).

Secondly, Waeyenbergh and Pintelon propose to identify the Most Important Systems (MISs; 7) within the manufacturing facility in order to be able to prioritise the systems. The identification of Most Important Systems is executed in order to find the systems and possibly issues which (can) influence the maintenance objectives most. Since many different objectives are considered when attempting to prioritise systems, a Multi-Criteria Decision Making technique is justified and therefore used in this study.

(20)

16 The third step is to zoom in on the systems and identify the Most Critical Components (MCCs) (8), searching for the components whose breakdown can affect or endanger the performance of the manufacturing system. For this, Waeyenbergh and Pintelon (2002) recommend the use of the Failure Mode Effect and Criticality Analysis (FMECA). This method is relatively quick and easy to use, while taking both quantitative and qualitative information into account. Based on the previous analyses, the ‘best’ maintenance policy can be selected for the given component, within the given system (9). In order to select the most appropriate maintenance policy, the decision-making tree of Waeyenbergh and Pintelon (2002) is used. However, this tool is adjusted according to the maintenance policy preferences of Scania Meppel, supported by research of Swanson (2001).

Next, Waeyenbergh and Pintelon (2002) state that if a preventive maintenance (PM) policy is chosen, its time-interval should be optimised (10). In addition, opportunistic maintenance (OM) is considered, creating a combined maintenance policy for a collection of components (Koochaki et al., 2011; Nicolai and Dekker, 2008). Koochaki et al. (2011) continue by arguing that opportunistic maintenance will help reducing maintenance costs and possibly improving availability. Other optimisations for maintenance policy are not discussed since these triggers are static and predefined.

The collection of maintenance policies of each component within each system represents the maintenance strategy (11). This step involves both planning and implementing the improved maintenance strategy. Next, since the maintenance strategy influences the performance of the manufacturing process (12), a performance analysis (13) should be made in order to evaluate the effectiveness of the developed policies through comparing performance targets (6). The performance analysis is used as a reference throughout the maintenance policy decision-making process. The performance originates from the current maintenance policy mix (maintenance strategy). Therefore, performance from previous periods serves as reference period for adjustments to the maintenance strategy. For this, a set of targets or Key Performance Indicators (KPI’s) need to be identified. These targets are derived from the alignment between the maintenance objectives and its resources. In appendix E, the selection of different KPI’s and targets are covered, based on De Groote et al. (1995) and Muchiri et al. (2010).

Finally, Márquez et al. (2009) add that the selected strategy should continuously evolve and consequently improve through the use of evaluation, again using both explicit and tacit knowledge. Therefore, the process elaborated in the paragraphs above is an iterative process, repeating itself regularly through a process of planning, implementing, evaluating, and improving the maintenance strategy, ensuring evolvement as recommended by Márquez et al. (2009).

(21)

17 5.1 Sub-conclusion

(22)

18

6. Maintenance policy decision-making framework application

In this chapter, the maintenance policy decision-making framework developed in chapter 5 (figure 5) is applied at Scania Meppel. First, in paragraph 6.1, decision steps 1-3 are covered, identifying the maintenance objectives at Scania. In addition, literature is consulted in order to identify possible additions and foundation. As stated, the maintenance objectives are derived from both corporate and manufacturing objectives. Thereafter, in paragraph 6.2, the resources and targets are identified in order to measure performance set against the maintenance objectives. This covers step 4 in the framework. Paragraph 6.3 covers the performance targets and the alignment of performance indicators and maintenance objectives, referring to decision steps 5 and 6. In the next paragraph, per objective, the performance is measured and corrected by the objective’s importance (weight), consequently identifying the most important systems, covering decision steps 7 and 13. Paragraph 6.5 covers decision step 8, where a specification of components is made, identifying its impact on their corresponding system. These decision steps ultimately lead to a classification of components. This forms input to the actual maintenance policy selection, and if applicable, preventive maintenance optimisation and opportunistic maintenance application, covering decision steps 9 and 10. Together, these steps form an improved maintenance strategy, aimed at improving the availability-cost trade-off.

6.1 Maintenance objectives of Scania

Muchiri et al. (2010) states that maintenance’s main objectives are related to availability, while quality-, factory operating- and safety conditions are met. From literature, different ‘main’ objectives are suggested. An overview of these different maintenance objectives is given in the table below. Note this list is non-exhaustive.

Bevilacqua et al. (2000)

Waeyenbergh and Pintelon (2002)

Muchiri et al. (2010) Wang et al. (2007)

Safety Safety Plant safety and

environment Safety: - Personnel; - Facilities; - Environment. Maintenance importance

for the process

Availability Plant functionality: - Availability; - Reliability; - Desired output (operate-ability); - Product quality. Costs: - Hardware; - Software; - Personnel training.

Failure frequency Longevity Ensuring plant achieves design life

Added-value:

- Spare parts inventories; - Production loss; - Fault identification. Downtime length Flexibility Cost effectiveness in

maintenance

Feasibility:

- Acceptance by labours; - Technique reliability. Operating conditions Performance Effective use of resources

Machine access difficulty Overtime Other (e.g. energy use) Spare part availability Costs (etc.)

(23)

19 From this overview, it can be shown that maintenance objectives are formulated on many different aggregative levels (e.g. strategic, tactical and operational). However, in order to identify the right maintenance objectives for Scania Meppel, corporate and manufacturing objectives need to be taken into account, as advocated by Tsang et al. (1999). In the following sections, the objectives of Scania and Scania Meppel are identified and, if applicable, complemented by supplementary objectives from literature (table 2).

The corporate objective of Scania is to “provide the best profitability for its customers throughout the Product life cycle by delivering optimised heavy trucks and buses, engines and services - thereby, becoming the leading company in its industry.” In order to achieve this, Scania incorporated core values throughout its organisation. These core values are customer satisfaction, respect for the individual, and quality, which are to be pursued by continuously improving the quality of products and services, efficiency (profitability and volume), job satisfaction and the elimination of all forms of waste (source: Scania.com). From the manufacturing perspective, Scania Meppel translated these objectives to:

- Customer first:

o Improve quality verification; o Decrease non direct runs; o Increase delivery precision. - Respect for the employee:

o Improve attendance percentage; o Improve safety;

o Improve employee satisfaction. - Elimination of waste:

o Increase Availability, Quality and Performance (OEE); o Increase direct runs;

o Reduce costs. (Source: Scania Inline, strategic platform)

In pursuit of these objectives, Scania Meppel strives to increase productivity, reliability and cycle time (average unit production time) through flexibility, while standardised methods and shared goals are attained for both production and supporting departments. These objectives are pursued using the Scania Production System, while taking the following priorities into account:

1. Safety, health, and environment; 2. Quality;

3. Delivery reliability;

4. Costs. (Source: Scania Inline, strategic platform)

(24)

20 al. (2011), who state that the main goal of a maintenance policy is to maximise plant availability and to minimise costs.

In the current situation, safety, health, and environment are met through the compliance to governmental and corporate rules and regulations regarding personnel, facility, and environment. This was also confirmed throughout the study by both maintenance technicians and paint shop engineers at Scania Meppel. This is also confirmed by the maintenance technicians at Scania Meppel. From the analysis of objectives at Scania, complemented by objectives from literature, the maintenance objectives are as follows:

1. Availability: a. Reliability:

O1. Failure frequency;

O2. Maintenance importance to process; O3. Quality of equipment maintenance. b. Maintainability:

O4. Spare part availability; O5. Personnel availability; O6. Machine access difficulty; O7. Downtime length;

2. Costs:

O8. Spare part inventory; O9. Personnel costs;

O10. Energy consumption per skid.

(25)

21 1. Availability 2. Costs Trade-off O8. Spare part inventory O9. Personnel costs O10. Energy consumption per skid A. Reliability B. Maintainability O1. Failure frequency O3. Quality of equipment maintenance O4. Spare part availability O5. Personnel availability O6. Machine access difficulty O7. Downtime length O2. Maintenance importance to process

Figure 6 - Maintenance objectives of Scania Meppel

6.2 Maintenance resources

The resources available to the maintenance department of Scania Meppel which are relevant for this study are the available man-hours and spare part inventory. Currently, 12 technicians, formerly 9, perform the maintenance activities within the production process. The available man-hours (12 fte.) are divided in two day shifts, Monday to Friday, and one nightshift, Monday to Thursday. During the day, mainly reactive maintenance is carried out, where in the night shift, proactive and aftermath of reactive maintenance is carried out. Roughly, the maintenance department has 18.000 man-hours per year or 1500 each.

(26)

22 6.3 Maintenance targets and indicators

Companywide, the OEE, appendix D, is used as a target indicator at Scania Meppel. For the OEE to reach a given level, for example the 82.5% goal for 2014, each indicator needs to be at least 82.5%. Maintenance targets of Scania Meppel are mainly focused on improving the availability of the production process. Current availability of the production process is 96% for the T1 line and 93% for the T2/T3 line. Still it is relevant to increase the availability since the remaining percentage of downtime disrupts the process and indirectly the quality of finished products (Muchiri et al. 2010). Furthermore, breakdowns and other outages itself and resolving them entail large amounts of resources.

Maintenance targets are to increase overall performance in the current situation, hence improve the current availability-cost trade-off. From this, the targets for the maintenance department are an improvement over the previous measurement of the performance of the maintenance strategy. Therefore, the maintenance strategy development is an iterative process of continuous improvement. Therefore, performance indicators were identified, specified in appendix E,in order to make a performance comparison of the maintenance targets and maintenance’s objectives.

According to Muchiri et al. (2010), performance indicators can be used in order to identify performance gaps between the current and desired performance objectives and can imply directions in order to close this performance gap. Also, De Groote (1995) states that performance indicators allow for, among others, taking immediate action, input for investment decision making and informing (third parties) of technical and economic progress. Furthermore, performance indicators are able to provide a link between strategy and operation and are therefore able to support continuous improvement of equipment’s performance (Muchiri, et al., 2010; Kaplan, 1983; White, 1996; Neely, 1999; Neely et al., 2005).

(27)

23 6.4 Most Important Systems

The listing below contains the systems identified by the maintenance department of Scania Meppel. For the physical location in the facility, refer to appendix K. Some systems as the Powerwash installation and the Robots & Ionisation can physically be separated from the other systems, whereas other systems as the conveyors and lifts and the software system are present throughout the system as a whole. The systems within the production process of Scania Meppel are specified as follows:

- Powerwash installation; - Paint washout system; - Conveyors and lifts; - Air treatment & RTO; - Pumproom 1; - Pumproom 2; - Robots & Ionisation; - Controlling software; - PDM and PLC;

- Manual Paint Line (MPL); - D-station (assembly).

In order to identify the importance of these systems, consequently prioritising each, the equipment performance, calculation steps elaborated in appendix E, of each system needs to be measured. In the table 3, the availability and costs over the period between 11-08-2014 and 31-10-2014 are given. The availability is based on data from SRMP failure registration system, whereas the costs are based on the value of the issued spare parts from the IFS information system (Scania Meppel). As shown in table 3, the availability of the Conveyors and Lifts is lowest, due to a relatively high frequency of short outages (low MTTF). These outages generally occur at loading, around interacting parts (e.g. doors), inspection and repair and unloading. Although downtime is short, outages especially at loading, can create disruptions throughout the whole production process. This is supported by Márquez et al. (2009) stating that variability earlier in the line will result in more disruptions throughout the process.

(28)

24 System Power-wash Paint washout Conv.& lifts Air tr. & RTO Pump room 1 Pump room 2 Robots & ionisation Controlling software PDM /PLC MPL D-station Frequency (failures) 3 16 352 23 37 164 255 3 5 7 27 MTTF(prod. hours) 687,29 435,6 13,9 246,72 64,25 35,17 63,48 196,32 438,81 294,87 84,13 MTTR(prod. hours) 0,63 0,06 0,29 0,77 0,32 0,16 0,73 0,26 0,45 0,58 0,41 Availability _99,91% _99,98% _97,95% _99,69% _99,50% _99,53% _98,86% _99,86% _99,9% _99,8% _99,51% Frequency (issued) 7 63 34 9 35 151 617 0 0 9 1 Costs (issued) € 625 € 7.716 € 1.418 € 1.936 € 2.884 € 46.970 € 126.468 € - € - € 311 € 35 % of total _0,33% _4,1% _0,75% _1,03% _1,53% _24,94% _67,14% _0% _0% _0,17% _0,02%

Table 3 - Availability and cost measures

Next to the indicators depicted in the table above, each maintenance objective of Scania Meppel, covered in paragraph 6.1, is measured for each system. In order to fully comprehend both performance and importance, the performance of each system is corrected by a, to be determined, weight. Thus, the performance regarding a given objective at a given system is measured and corrected by the assigned weight (importance). Consequently, the summation of each corrected value of each objective at a system will yield the classification of systems. Furthermore, the ratio between the availability and costs per system is identified. From this and due to the qualitative and intangible nature of the ‘importance’ of each objective, the Fuzzy Analytical Hierarchical Process (Wang et al., 2007) is executed in the paragraph below.

6.4.1 Fuzzy Analytical Hierarchical Process

The FAHP proposed by Wang et al. (2007) is composed of a number of methodological steps. First the goal or problem is hierarchically organised, followed by the creation of judgement matrices using pair-wise comparison. Next, following the identification of importance (priority) in relation to each other, the objectives are weighted. Finally each alternative (element) is ranked according to its priority. However, the identification of importance per system at Scania Meppel, performance measurement per objective and the ratio calculation of priority, through the identified weights, and performance are added. This way, qualitative (objective priority allocation) and quantitative (performance measures) data are combined, consequently identifying the Most Important System ranking for the production process of Scania Meppel. The steps used in this FAHP are therefore as follows:

1. Hierarchical organisation of the goal;

2. Create judgement matrices by pair-wise comparison; 3. Weight calculation per objective from matrices; 4. Performance measurement per objective; 5. Ratio calculation priority and performance; 6. System ranking according to priority.

(29)

25 importance of each system, the maintenance objectives are quantified, through performance indicators, for each system and weighted in relation to each other. The corresponding weights are identified using fuzzy judgement scores (Wang et al., 2007) depicted in appendix G. In figure 7, the hierarchical structure of the maintenance objectives is depicted

1. Availability Reliability Maintainability Failure frequency Maintenance importance to process Quality of equipment maintenance Spare part availability Personnel availability Machine access difficulty Downtime length Spare part inventories Personnel costs Energy consumption per skid 2. Costs System importance Objective importance

Figure 7 - Hierarchical structure of the fuzzy analytical hierarchical process

The next step in the FAHP is the development of judgement matrices by pair-wise comparisons. These matrices depict the judgement of each objective as compared to each other by experts of Scania Meppel. For this, fuzzy judgement scores are used as proposed by Ayhan (2013). These judgement score use a range, the triangular fuzzy numbers (TFN’s) consisting of a lower ( ), mean ( ) and upper ( ) bound (e.g. strongly important ( ) = (6, 7, 8); appendix F) comparing each objective combination ( ) as shown in the matrix below. As shown in the matrix below, each judgement score has an inverse (e.g. has inverse ).

A =

Table 4 - Fuzzy pair-wise comparison matrix setup

(30)

26 Table 5 – Fuzzy pair-wise comparison matrix of maintenance objectives of Scania Meppel

(31)

27 In order to calculate the priorities (weights) of each objective, Chang’s extent analysis is used (Balli & Korukoğlu, 2009). For this, the calculations 1 to 5 depicted and elaborated in appendix F were executed. Using the input from depicted in table 3, the resulting normalised weights ( ) are as follows:

Maintenance objective Indicator Weight

( )

Maintenance importance to process O2 0,211

Spare part availability O4 0,169

Downtime length O7 0,157

Personnel availability O5 0,154

Failure frequency O1 0,134

Quality of equipment maintenance O3 0,122

Spare part inventory O8 0,045

Machine access difficulty O6 0,000

Personnel costs O9 0,009

Energy consumption per skid O10 0,000

Total 1

Table 6 - Normalised maintenance objective weights

From table 6, it can be concluded that the first six objectives are considered the most significant, emphasising the maintenance importance to process (0,211). The maintenance importance to process shows the influence of a system or component has (criticality) on the performance of the production process as a whole, hence the importance of maintenance activities on a system or component. In other words, the consequences of breakdowns are regarded as most important and should be controlled, minimised, or avoided altogether.

Now that the weights have been identified, the next step is to adjust the normalised performance measures per objective to their corresponding weight. This way, both importance and performance of each system and objective therein are taken into account, creating a combined and interrelated measure (impact indicator). This measure therefore incorporates both the interdependency of systems and the interdependency of maintenance objectives.

6.4.2 Maintenance’s weighted performance

(32)

28 Table 7 – Normalised subdivision of performance per system (rounded)

System Failure frequency Maintenance importance to process Quality of equipment maintenance Spare part availability Personnel availability Machine access difficulty Down-time length Spare part inventory Personnel costs Energy consumption per skid Powerwash (PW) 0,003 0,000 0,001 0,030 0,030 - 0,005 0,003 0,00 0,223 Paint washout system (WWT) 0,018 0,000 0,009 0,061 0,061 0,222 0,003 0,041 0,03 0,026 Conveyors & lifts 0,395 0,263 0,234 0,152 0,152 0,167 0,276 0,008 0,17 0,050

Air treatment &

(33)

29 In table 7, the mutual relationship between systems is given by measuring and normalising the performance per objective. However, this does not yet accurately pinpoint the objectives and systems who need increased attention, hence are the Most Important Systems. In order to create a representative image, including both importance (significance) of objectives and performance of each system, the performance per system from table 7 is corrected by their corresponding FAHP weight (importance). This creates an interrelated impact indicator of both system’s performance and objectives. For example, the failure frequency (performance) at the Conveyors and Lifts system (0.3946 ≈ 0.395) is corrected by its allocated importance (FAHP score; table 6) of 0.1337 (≈ 0.134), creating a performance indicator of 0.5275 (≈ 0.523). In short, the maintenance weighted performance per objective, per system is the product of objective importance and system performance. These impact indicators of the actual performance and objective importance allocation is depicted on the next page, in table 8.

(34)

30 Table 8 – Maintenance weighted performance per objective, per system

Maintenance objective FAHP

Score Powerwash (PW) Paint washout system (WWT) Conv. & lifts Air treatment & RTO Pump-room 1 Pump-room 2 Robots & ionisation Controlli ng software PDM/ PLC MPL D-station Total

O1. Failure frequency 13,37% 0,04% 0,24% 5,28% 0,34% 0,55% 2,46% 3,82% 0,04% 0,07% 0,10% 0,40% 13%

O2. Maintenance

importance to process 21,11% 0,00% 0,00% 5,54% 0,06% 0,05% 0,46% 14,98% 0,00% 0,00% 0,00% 0,02% 21%

O3. Quality of equipment

maintenance 12,18% 0,02% 0,11% 2,85% 0,27% 2,03% 1,31% 4,28% 0,06% 0,04% 0,08% 1,12% 12%

O4. Spare part availability 16,91% 2,56% 2,05% 0,51% 1,02% 1,28% 0,77% 0,26% 2,82% 2,31% 1,79% 1,54% 17%

O5. Personnel availability 15,38% 2,33% 1,86% 0,47% 0,93% 1,17% 0,70% 0,23% 2,56% 2,10% 1,63% 1,40% 15%

C6. Machine access

difficulty 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0%

O7. Downtime length 15,66% 0,08% 0,04% 4,32% 0,75% 0,50% 1,14% 8,06% 0,03% 0,10% 0,17% 0,47% 16%

O8. Spare part inventory 4,51% 0,01% 0,18% 0,03% 0,05% 0,07% 1,12% 3,03% 0,00% 0,00% 0,01% 0,00% 5%

O9. Personnel costs 0,89% 0,00% 0,03% 0,16% 0,12% 0,05% 0,12% 0,37% 0,00% 0,00% 0,01% 0,02% 1%

O10. Energy consumption

per skid 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0,00% 0%

(35)

31 6.5 Sub-conclusion

In the previous paragraphs, 6.1 – 6.4, the elements in play and its demarcation of the maintenance function of Scania Meppel were retrieved. First, from both literature and Scania Meppel, the maintenance objectives were identified, which together form the availability-cost trade-off. Next maintenance’s resources, performance targets, and indicators. From the FAHP, the objectives were prioritised using a weight. The highest scoring objectives are the ‘maintenance importance to process’ (.211), ‘spare part availability’ (.169), and ‘downtime length’ (.157) are all focused on avoiding, minimising, or controlling downtime and its consequences. Next, from combining both performance and objective importance into an impact indicator, the Most Important Systems were identified. From this, the Robots and Ionisation, Conveyors and Lifts, and Pumproom 2 are to be emphasised when maintenance decisions are made.

6.6 Most Important Component analysis

In the previous steps of the maintenance policy decision-making framework, the maintenance objectives were identified and quantified according to their importance. Next, the systems which apply to the maintenance function of Scania Meppel were identified. Next, by calculations, the Most Important Systems (MISs) were through the use of impact indicators, composed of actual performance and importance (weight) of each maintenance objective. The next step in the framework is to identify the components within the systems. The table below depicts the components within each system, where the systems are ranked on importance. This listing has been established through data from the SRMP failure registration system, supplemented by the technicians of the maintenance department.

Power-wash installation Paint washout system Conv. and lifts Air treatment & RTO Pumproom 1 Pumproom 2 Robots & Ionisation Controlling software MPL D-station PDM & PLC PW1 PW2 Coagulati on Decanter Pump Skimmer ASU Gas RTO Cooling system OM## Software T1 T2/T3 All lines Robot mechanical Robot electrical Gas system Air system Ignition system T1/P1/ WBC/ CC/T3 Flaming High-tension system Nozzles EPP station Robot mechanical Robot electrical Harder system Colour change block Gunbox WRSB Blank Cabin T2 PDM

Figure 8 – Components within the systems at Scania Meppel

(36)

32 Subsequently, in the next paragraph, the performance of each component is reviewed, identifying the criticality or importance of each component within a system. For this, Waeyenbergh & Pintelon (2002) proposed a Failure Mode Effect and Criticality Analysis, executed in the next paragraph.

6.6.1 Failure Mode Effect and Criticality Analysis

The FMECA, proposed by Waeyenbergh and Pintelon (2002) enables a judgement whether a system’s component should be prioritised. In addition, this analysis yields practical implications regarding the selection of a maintenance policy, as for example the failure frequency and condition measurement. This judgement is based on both likelihood and severity. The FMECA, as shown in appendix H, is tailored specifically for Scania Meppel in order to identify the Most Important Components within the Most Important Systems.

Due to time limitations of this study, the three Most Important Systems are analysed using the FMECA. These are, as identified in the previous paragraph, the Robots and Ionisation, Conveyors and Lifts, and Pumproom 2. Overall, both technically and economically (table 3), these systems prove to be the largest malefactors to the equipment performance of the production process as a whole. In the table below, a selection of the results from the FMECA of the T1/P1/WBC/CC/T3 components of the Robots and Ionisation system is shown. First, the costs of the work orders from between 11-08-2014 to 31-10-11-08-2014 are given, followed by the failure modes per component, their criticality score, followed by some detail from the analysis as the mean time to failure (MTTF) and means of measuring the condition of the component concerned.

Maintenance strategy development