TRADE-OFF FOR SUPPLY CHAIN RESILIENCE: THE EFFECTS OF EFFICIENCY AND REDUNDANCY IN MAINTENANCE

(1)

THE EFFECTS OF EFFICIENCY AND REDUNDANCY IN MAINTENANCE

Master’s Thesis Supply Chain Management, MSc

University of Groningen, Faculty of Economics and Business

STEPHANIE BRÄUER

Student number: S2757222

E-mail: s.braeuer.1@student.rug.nl

Supervisor, University of Groningen: Dr. Kirstin Scholten

Co-Assessor, University of Groningen: Prof. Dr. Dirk-Pieter van Donk

January 25th, 2016

Acknowledgments:

(2)

THE EFFECTS OF EFFICIENCY AND REDUNDANCY IN MAINTENANCE

Abstract

Purpose – This paper aims to explore how to design resource structures in maintenance to balance efficiency and redundancy for supply network resilience. Specific resource types and their linked key decisions in relation to the phases prior and after a supply network disruption to enable an effective and efficient response are investigated.

Design / Methodology / Approach – A mixed method consisting of an explorative case study of a single company’s supply network and a design science methodology was conducted. Data were gathered from 30 interviews with 15 different interviewees, a questionnaire as well as archival information.

Findings –Twelve mechanisms were identified that reveal the significant trade-offs regarding efficiency and redundancy between information, manpower, spare parts and tools in maintenance necessary in order to design resilience into the supply network.

Originality / Value – This is one of the first empirical investigations to provide deeper insights into the topic of supply chain resilience linked to maintenance, in particular into the trade-off between efficiency and redundancy of maintenance resources aimed at achieving supply chain resilience. The developed decision making model offers understanding into the mentioned trade-off and is therefore of theoretical and practical importance.

(3)

3

CONTENT

LIST OF FIGURES ... 4 LIST OF TABLES ... 4 1 INTRODUCTION ... 5 2 THEORETICAL BACKGROUND ... 6

2.1 Supply Network Resilience ... 6

2.2 Maintenance Resource Design...10

2.3 Conceptual Model ...13 3 METHODOLOGY ...14 3.1 Research Context ...16 3.2 Scenario Selection ...17 3.3 Data Collection ...20 3.4 Data Analysis ...22 4 FINDINGS ...24

4.1 Findings regarding Information ...24

4.2 Findings regarding Manpower ...26

4.3 Findings regarding Spare parts ...28

4.4 Findings regarding Tools ...29

4.5 Overall Findings ...30

5 DISCUSSION ...35

5.1 Resource Balancing in the Preparation ...37

5.2 Resource Balancing in the Response ...38

5.3 Resource Balancing in the Recovery ...38

6 CONCLUSION ...39

6.1 Managerial Implications ...40

6.2 Limitations and Further Research ...41

LIST OF REFERENCES ...42

APPENDIX ...45

Appendix A: Case Study Protocol ...45

Appendix B: Interview Questions ...47

Appendix C: Questionnaire / Supply Chain Mapping of Scenarios ...49

Appendix D: Supply Chain Mapping of As Is Situation ...54

(4)

4

LIST OF FIGURES

FIGURE 2-1: Supply Chain Disruption Profile ... 7

FIGURE 2-2: Conceptual Model ...13

FIGURE 3-1: Overall Research Design Process ...15

FIGURE 3-2: Link between Practice and Theory for SCRES ...16

FIGURE 3-3: Objectives, Constraints and Variables of the Problem Definition ...17

FIGURE 3-4: Coding Tree ...22

FIGURE 4-1: Scenario Comparison of Quantitative Findings: Efficiency & Redundancy ...31

FIGURE 4-2: Scenario Comparison of Quantitative Findings: SCRES Phases ...31

FIGURE 5-1: SCRES Decision Making Model ...36

LIST OF TABLES

TABLE 2-1: SCRES Phases and their Capabilities regarding Effectiveness and Efficiency ... 9

TABLE 2-2: Maintenance Resource Types and their related Structuring Decisions ...12

TABLE 3-1: Scenario Selection Overview ...18

TABLE 3-2: Overview of Data Collection ...21

TABLE 3-3: Overview of Codes and assigned Resource Types ...23

TABLE 3-4: Ensuring Data Quality ...23

TABLE 4-1: Scenario Comparison of Qualitative Findings ...32

TABLE 4-2: Mechanisms, their Dependencies and Linkage to the Resource Types ...33

(5)

5

1 INTRODUCTION

Supply chain resilience (SCRES) is the capability of simultaneously enabling an efficient as well as effective response to unforeseeable disruptive events in the supply network (Christopher & Peck, 2004; Ponomarov & Holcomb, 2009; Scholten, Sharkey, & Fynes, 2014). Maintenance is seen as a facilitator of SCRES as it ensures the reliability of systems (Saleh & Marais, 2006) and thus, the ability of continuous performance (Söderholm, Holmgren, & Klefsjö, 2007). It is facilitated by the appropriate design of resource employment which involves, for example, decisions regarding the composition of mechanics or location of equipment (Kelly, 2006). Redundancies of resources, such as surplus information or spare parts inventory in maintenance, that help to prepare for, respond to and recover from disruptions, are seen as an enabler of SCRES (Sheffi & Rice Jr., 2005). At the same time, however, supply networks including maintenance are often designed to optimize cost and efficiency (e.g. the waste reduction focus of lean) (Marais & Saleh, 2009; Marais, 2013; Rosqvist, Laakso, & Reunanen, 2009). As strategies based around lowest cost aim to take away possible resource buffers which can hinder maintenance performance and thereby a desired operational condition of systems (Kelly, 2006; Rosqvist et al., 2009), it might impact the network’s vulnerability and hence SCRES (Blackhurst, Dunn, & Craighead, 2011; Ivanov, Sokolov, & Dolgui, 2014). To manage the strategic fit between efficiency and effectiveness of SCRES, the appropriate employment of maintenance resources regarding efficiency and redundancy is important to understand and thus, focus of this study.

The mentioned issue is underpinned by the disagreement of different authors on this topic. Christopher & Peck (2004), for example, argue that if a company decides to become more resilient, e.g. by meeting high maintenance reliability requirements (Saleh & Marais, 2006), the strategic surplus resources have to be chosen over short term cost optimisation (Christopher & Peck, 2004; Zsidisin & Wagner, 2010). Otherwise, pure cost efficiency will come with even higher cost should disruptions occur as there are no resources available for an effective response (Tang, 2006) such as the repair of failed systems. However, the benefits of using resource redundancies for reducing the vulnerability of the supply network are limited (Tang, 2006; Zsidisin & Wagner, 2010). At some point the employment of additional mechanics or increase of repair material will only result in a decreased gain for maintenance reliability (Saleh & Marais, 2006) and thus, SCRES, since resource investments will absorb the financial profits (Brandon-Jones, Squire, Autry, & Petersen, 2014).

(6)

6 efficiency and redundancy for supply network resilience. In answering the research question, this study makes key contributions to theory and practice. This research is of theoretical relevance, because it provides further insights into the topic of SCRES based on empirical investigation by linking SCRES to maintenance. In particular, this study develops an understanding of the included trade-off between efficiency and redundancy of resources in maintenance design. Further, this study offers support to guide companies in their decision making towards resilient, i.e. effective and efficient, supply networks by giving managerial advice for the design of maintenance in terms of resource structure.

This paper is structured as follows. A review of the existing literature in relation to supply network resilience and maintenance design of resources will mark the beginning. The potential interaction will then be clarified through the construction of a conceptual model. This will be followed by a description of and argumentation for the chosen methodology. The outcomes of the data collection and analysis with regards to the conceptual model will be presented. Further, a discussion will elaborate the representativeness and applicability of the model. A final section with the aim of answering the research question and providing managerial implications and limitations as well as further research indications will conclude.

2 THEORETICAL BACKGROUND

2.1 Supply Network Resilience

SC disruptions are connected to any unanticipated and inevitable event (Blackhurst et al., 2011; Skipper & Hanna, 2009) that impacts the normal flow of goods, material and / or services (Craighead, Blackhurst, Rungtusanatham, & Handfield, 2007) causing significant financial and operational losses and are therefore main challenges of SCM (Sheffi & Rice Jr., 2005; Simchi-Levi & Wei, 2012). Ponomarov & Holcomb (2009:131) define SCRES as “the adaptive

capability of the supply chain to prepare for unexpected events, respond to disruptions, and recover from them by maintaining continuity of operations at the desired level of connectedness and control over structure and function.” Following this definition SCRES can

(7)

7 FIGURE 2-1: Supply Chain Disruption Profile

Although neither the kind, time or place of a disruption can be fully anticipated, the needs of the organization during an possible event can (Hopp, 2008). Thus, preparation refers to the ability to control and monitor risks (Ivanov et al., 2014) and thereby foresee and get ready in the operational design in order to minimize negative effects (Sheffi & Rice Jr., 2005). The response phase is devoted to processes of how to react in the case of a disturbance (Ivanov et al., 2014) and develops a plan and responsive adjustment in real time while the event is unfolding (Hopp, 2008). It is described as the time between the occurrence of a disruption and the realization of its full intensity (Sheffi & Rice Jr., 2005). The overall disruption impact (shaded area in figure 2-1), depends on the disruption intensity (severity of a disruption) and the according disruption time which are both influenced by the nature of the event and the characteristics of the company’s response (Sheffi & Rice Jr., 2005; Tukamuhabwa et al., 2015). In the response phase the performance loss can still be influenced (Jüttner & Maklan, 2011; Skipper & Hanna, 2009) by reducing its impact intensity and / or its timely length (Sheffi & Rice Jr., 2005; Tukamuhabwa et al., 2015). The recovery includes processes for adaptation (Ivanov et al., 2014) and concerns the time between the realization of the full intensity of the disruption and recreation of a stable performance level. Although the maximum performance impact cannot be affected anymore, the long lasting negative effect can be lowered over time (Sheffi & Rice Jr., 2005). In the end, the recovery can result in either a lower (shortfall) or a higher (growth) performance level than prior the disruption (Sheffi & Rice Jr., 2005; Tukamuhabwa et al., 2015). Although divided into the named three phases, SCRES is a holistic concept and an ongoing process as every recovery is followed by a preparation for the next disruption.

(8)

8 characterized by efficiency (Christopher & Peck, 2004; Jüttner & Maklan, 2011; Stevenson & Spring, 2007). Nevertheless, capabilities are interrelated and later phases are influenced by decisions made in earlier ones. Table 2-1 shows which particular supply chain capabilities are important for efficiency and effectiveness in the different phases. To allow the supply network to resist change without adapting its initial stable configuration (Wieland & Wallenburg, 2012) additional physical as well as information resources have to be secured already in the preparation phase (Bhamra, Dani, & Burnard, 2011; Blackhurst et al., 2011; Timothy J. Pettit & Croxton, 2010). These resource buffers either delay the impact of a disruption and thus, provide more time to plan, or enable a faster response directly (Zsidisin & Wagner, 2010). These needed resources have to be actually hold in readiness or virtually secured; the first will enable a fast but costly response while the latter will be a cheaper and slower option (Hopp, 2008). The availability of resources contributes to a faster response, whereas the built-in flexibility contains an effective bolstering against disruptions (Sheffi & Rice Jr., 2005; Skipper & Hanna, 2009; Zsidisin & Wagner, 2010) as resources can easily be employed, redeployed or redirected (T.J. Pettit, Croxton, & Fiksel, 2013; Sheffi & Rice Jr., 2005) to encounter, resolve and exploit threats (Jüttner & Maklan, 2011).

(9)

9 TABLE 2-1: SCRES Phases and their Capabilities regarding Effectiveness and Efficiency

EFFECTIVENESS EFFICIENCY P rep ar atio n Robustness Robustness

- Anticipation of disruptions by e.g. monitoring, forecasting or contingency planning (Bhamra et al., 2011; Blackhurst et al., 2011; Timothy J. Pettit & Croxton, 2010; Wieland & Wallenburg, 2013)

- Availability of information or physical resources enabled by SC collaboration (Bakshi & Kleindorfer, 2009; Tang, 2006; Wieland & Wallenburg, 2013)

- Information about entities, assets and events in the whole supply network (T.J. Pettit et al., 2013; Wei & Wang, 2010; Wieland & Wallenburg, 2013) enhance an effective response and recovery (Jüttner & Maklan, 2011)

- Information resources enhance the capability of efficient coordination of operations (Tang, 2006) since a lack of shared information is source of additional cost (Christopher & Peck, 2004)

Flexibility

- Number of possible states a supply chain can take (Stevenson & Spring, 2007)

- Flexibility buffer or contingent resources can serve multiple purposes (Hopp, 2008; Tukamuhabwa et al., 2015).

Respo

n

se

Flexibility Velocity

- Redirection employment and redeployment of resources (T.J. Pettit et al., 2013; Sheffi & Rice Jr., 2005)

- Availability of alternative choices / redundancies (Jüttner & Maklan, 2011; Sheffi & Rice Jr., 2005)

- Pace of detection (Manuj & Mentzer, 2008) and flexible adaptions (Stevenson & Spring, 2007)

- Determines the occurring loss per unit of time (Jüttner & Maklan, 2011)

Recove

ry

Flexibility Velocity

- Redirection or employment and redeployment of resources (T.J. Pettit et al., 2013; Sheffi & Rice Jr., 2005)

- Availability of alternative choices (Sheffi & Rice Jr., 2005)

(10)

10

2.2 Maintenance Resource Design

Since also infrastructure systems are not economically designed for all eventualities, SC disruptions can result in breakdowns of components and thus, shortages and interruptions of the usability of the systems (Balzer & Schorn, 2015). Due to companies’ dependability on technical systems and its high financial values, maintenance gained a huge importance in supply networks in relation to disturbances as it reduces business risks by hazard elimination and ensures the ability of continuous performance (Söderholm et al., 2007) and thus, avoids costs of failure (Marais & Saleh, 2009). Maintenance involves all technical and administrative actions (Dekker, 1996) intended to secure (preventive, proactive to a failure) or restore (corrective, reactive to a failure) a system and its assets in or to a desired operational condition in which they can perform the required functions and thereby ensure reliability (Balzer & Schorn, 2015; Marais & Saleh, 2009; Pham & Wang, 1996; Rosqvist et al., 2009). This requires the appropriate allocation and use of resources which can be a significant driver of competitiveness and thus, source of value (Balzer & Schorn, 2015; Marais & Saleh, 2009). Marais & Saleh (2009) advise to consider the cost of maintenance and the resulting system reliability together when developing optimal maintenance strategies. The aim of any resource structure design is a maximum resource performance and efficiency achieved by high resource utilization (low costs) for a desired speed of response and quality of work (Kelly, 2006). Although redundancies in resources promote high performance they also cause high costs. Besides the direct cost of these redundancies also e.g. the short-term downtime or loss of productivity resulting from preventive maintenance are costly (Saleh & Marais, 2006). Here, a variety of physical, financial, human and information resources are required (Hastings, 2015) to meet the maintenance workload at a given plant layout and technological sophistications (Kelly, 2006). The challenge for maintenance is therefore to identify appropriate objects and tasks and ensure the availability of sufficient resources (Rosqvist et al., 2009) involving decisions about the redundant and efficient employment of information, manpower, spare parts and tools as defined by Kelly (2006). Table 2-2 shows these different resource types and their related structuring decisions. The shown decision can be interrelated with decisions of other resources and structuring therefore has to be seen as a whole picture (Kelly, 2006).

Information includes all documents, catalogs, manuals or drawings that might facilitate maintenance work under the categories of training, reference (e.g. manuals, item histories, spares catalogs), instruction (e.g. work orders, compulsory job instructions, safety instructions), scheduling (e.g. preventive routines) and control (Hastings, 2015; Kelly, 2006).

(11)

11 specialization (e.g. boiler, fitter, turbine fitter) (Kelly, 2006). Additionally to tradesmen also other categories of staff like planners or store men have to be considered (Hastings, 2015). Spare parts are consumable items that have a significant influence on the operability, stability and costs within an infrastructure system (Balzer & Schorn, 2015) as they service internal assets or customer machinery (The Logistics & Supply Chain Management Society, 2015) and thus, sustain a certain operation (Kelly, 2006). They can be classified by their specific function (e.g. abrasives, bearings) (Kelly, 2006). The main challenge lies in the variety and complexity of the many different items (varying cost, lead times and usage rates) (Kelly, 2006). Tools are non-consumable items used to perform or facilitate work (Balzer & Schorn, 2015; Kelly, 2006). Here, the main challenge lies in the monitoring of their loan and maintaining them (or replacing them if necessary) when returned (Kelly, 2006).

(12)

12 TABLE 2-2: Maintenance Resource Types and their related Structuring Decisions

Maintenance Resource Type

Key decisions

Description

Manpower Composition - Ensuring sufficient competence to perform defined tasks (Balzer & Schorn, 2015) - Defining work roles, skills and degree of flexibility (Hastings, 2015; Kelly, 2006) - Knowledge and high productivity via specialization / work division (Kelly, 2006)

- Cross-trained workers for higher utilization and flexibility (Hastings, 2015; Hopp, 2008; Kelly, 2006)

- Consideration of external personnel for specialized work or demand peaks (Balzer & Schorn, 2015; Kelly, 2006) Location

allocation

- Dedication to the maintenance of a single plant, area or unit type (Hastings, 2015; Kelly, 2006)

- Specific knowledge and quality (decentralization) versus flexibility and utilization (centralization) (Kelly, 2006) Sizing - Ensuring sufficient capacity to perform the defined tasks (Balzer & Schorn, 2015; Hastings, 2015)

- Shift working during day, night and weekends (Kelly, 2006) - Cost of unavailability versus labour utilization (Kelly, 2006) Logistics - Management of the movement of employees (Kelly, 2006)

Spare parts Sizing - Quantity of stock (Balzer & Schorn, 2015; Hopp, 2008)

- Ordering and holding costs versus stock out costs (Kelly, 2006)

Sourcing - Covering demand by selecting one or several suppliers (Balzer & Schorn, 2015; Hastings, 2015) - Influence of lead times on inventories (Kelly, 2006)

Location - Function, location and number of stores (Hastings, 2015; Kelly, 2006)

- Central storage with less stock versus decentralized storage with smaller time-wise distance (Hopp, 2008) Logistics - Management of the movement of spares (Kelly, 2006)

Tools Location - Identifying an optimal or effective positioning (Kelly, 2006)

- Dependent on size and frequency of use (e.g. supplied to individual artisans or held for issue, in a tool stores) (Kelly, 2006) Logistics - Management of the movement of tools (Kelly, 2006)

Information Purpose - Defined users of information (Kelly, 2006)

Form - Way of holding information (e.g. paper, computer, combination) (Kelly, 2006) Location - Identification of an optimal or, at least, effective positioning (Kelly, 2006)

- Centrally, locally or combination (Kelly, 2006)

(13)

13

2.3 Conceptual Model

The aim of this study is to understand how to design resources in maintenance to balance efficiency and redundancy for supply network resilience. SCRES was defined by its process phases of preparation, response and recovery. Maintenance resource design was subsequently classified by its underlying resource types (manpower, spare parts, tools and information) and operationalized by their structuring decisions. These maintenance resources, deployed in redundancy and efficiency, contribute to increased reliability of the supply network and thereby indicate ways in which maintenance, containing both proactive and reactive elements, has influence on each of the named phases and thereby on SCRES itself. Aligning maintenance resource design with the study of resilience suggests that maintenance can aid resilience by providing the needed resources while focusing on redundancy and efficiency to successfully manage the strategic fit for the different phases. Based on these conceptualizations the theoretical model underlying this research has been developed and will guide the research. Figure 2-2 shows the conceptual model which depicts the linkages between maintenance resources, employed in redundancy or efficiency, the disruption phases and finally SCRES. The model reflects theory further as it suggests a bigger share of redundancy in the beginning and efficiency towards the end of the disruption profile as explained in section 2.1.

(14)

14

3 METHODOLOGY

In order to answer the underlying research question and investigate the constructed conceptual model, an empirical study is opted for by applying a design science methodology as part of a single case study research. This mixed methods research is beneficial in answering the complex research question and collect rich data (Yin, 2009). To generate meaningful, relevant theory gained from the internal logical understanding out of the explorative empirical investigation, a case study design is the most suitable method (Eisenhardt & Graebner, 2007; Eisenhardt, 1989). With this method the specific and complex phenomenon of SCRES and its intertwining with the supply network can be explored in-depth in a real-life context (Yin, 2009). Since this research is explorative in nature and complex by means of its multiple variables (operationalization shown in table 2-1 and 2-2) and different levels, it requires the flexible process and rich data collection a case study design offers (Voss, 2009). To support this latter need further, studying a single firm as one case enables the gathering of more detailed information due to in depth observation (Voss, 2009). The unit of analysis of this study, derived from the research question, is a supply network. This single case of a supply network further offers a practical heretofore unsolved problem (Hevner, March, Park, & Ram, 2004) that will be investigated using design science to find an applicable solution for it (Hevner et al., 2004; Holmström, Ketokivi, & Hameri, 2009). The aim of this study is not only to develop explanations but shaping the phenomenon of interest (Simon, 2002) and thereby improving practice (Simon, 1996). This study distinguishes three interlinked phases in the design process, “framing the problem and solution objective”, “design development” and “solution refinement1_{”, which are} all connecting the business environment, the research and the knowledge base (Hevner, 2007; Peffers, Tuunanen, Rothenberger, & Chatterjee, 2007). Figure 3-1 shows the different steps and their output respectively input during the research. Section three of this paper will follow the theoretical set-up which has been established using inputs from the practical steps.

1_{Since in the step of solution refinement no further insights of theoretical relevance were encountered, this phase solely serves}

(15)

(16)

16

3.1 Research Context

In order to study the phenomenon of SCRES, the Waterbedrijf Groningen, a public water utilization company, was chosen as the overall research context. Since water is a staple and the main contributor of hygiene and thus, resulting in health and living standard, the management of risk has a high significance for the Waterbedrijf as the kind of service implies an ethical angle. The ability to supply water is measured in downtime experienced by customers which can either occur due to planned activities or unplanned damages of the pipe system. This service level is achieved by operational performance, whereas low price rates, which are awaited by public authorities and customers, are connected with operational costs. In this way the focal company is most suitable for this research as it seeks for both effectiveness and efficiency and thus, to increase its supply network’s resilience as illustrated in figure 3-2. Moreover, maintenance holds a great importance for the company as it ensures the continuous performance of its physical assets in its very spacious and relative inaccessible, underground, distribution network. Lastly, the organization has an interest in understanding the trade-off effects of its resource structure in maintenance, which are so far unexplored. SCRES in the Waterbedrijf can be measured as the response time to a disruption. This is the time from the moment a disruption is detected to the moment it is solved (i.e. a pipe breakdown is solved). The response time is influenced by the design of the maintenance process of the supply network.

FIGURE 3-2: Link between Practice and Theory for SCRES

(17)

17 which will replace the current situation with one central warehouse and two satellite warehouses (see appendix D for the supply chain map of the as is situation). However, the new overall maintenance process is not finalized. Within the new layout there exist many design options. Although the importance of maintenance as a central activity as well as for building resilience is consistent throughout the company, various stakeholders hold different priorities and desired outcomes regarding the design of the maintenance process. Thus, it is unclear how to design the current maintenance network. Based on the information given by the company, figure 3-3 displays the objectives, constraints and variables of the problem of designing the maintenance process.

FIGURE 3-3: Objectives, Constraints and Variables of the Problem Definition

3.2 Scenario Selection

(18)

18 TABLE 3-1: Scenario Selection Overview

Scenarios

Maintenance Resources

Information Manpower Spare parts Tools

As is situation

(base scenario)

Central warehouse:

- MRP system: Stock level based on parameters; Consideration of delivery times

- Little visibility into suppliers’ stock

- 4 storemen

- Incl. 1 driver (1 truck) on unofficial call-on duty at night / weekends

- Bigger and more specialized spare parts

- Bigger and more specialized tools

Satellite warehouses (2):

-MRP system linked to central warehouse: No parameters for inventory policy

- Paper note about taken material (not completely implemented); Stocktake every week

- Planners create weekly schedules

- 1 storeman each, 1 driver + truck for both locations

- Small and medium spare parts

- Small and medium tools

Fitters’ trucks:

- No information about stock

Fitters:

- 2 teams each 15 FTE

- One truck per fitters (a lot of driving) - 4 fitters on call-on duty at night / weekends

- Small “every day” spare parts

- Small “every day” tools

Contractors

- Diverse agreements without visibility into resources

Contractors:

- Machine operators (24/7)

- Mechanics at day time during week

Contractors:

- Small “every day” spare parts (day time during week)

Contractors:

- Big machinery at day / night / weekends

Small cabins

New central warehouse:

- Existing MRP System

- More visibility into suppliers’ stock - Planners will create monthly schedules shared with store men - Coordination of delivery

- Direct delivery for disruptions and planned work requires more drivers (competence and amount) and trucks (24h) - Reduction or shift of satellite employees

- All spare parts (reduction due to centralization)

- Bigger and more specialized tools

Small cabins:

- Automatic inventory system

Small cabins:

- New work roles needed due to administration demand

Small cabins:

- No material planned

Small cabins:

- Small and medium tools (overall less tools because of no satellites)

- Agreement on parameters - Automatic inventory system linked to the new central warehouse

Fitters:

- 1 team; each fitter one truck

(reduction of fitters conceivable due to centralization; Less driving more core work) - 4 fitter on call-on duty at night / weekends

Contractors

- Diverse agreements without visibility into resources

Contractors:

- Mechanics (day time during week)

Contractors:

(19)

19

CW pick-up

- Existing MRP system

- More visibility into suppliers’ stock - Planners will create monthly schedules shared with store men - Coordination of pick-ups

- Direct delivery for disruptions requires more drivers (change in competence not necessarily amount) and trucks (24h) - Reduction or shift of satellite employees

- All kind of tools (reduction due to centralization)

Fitters:

- 1 team; each fitter one truck (reduction of fitters conceivable)

- 4 fitter on call-on duty at night / weekends

Contractors

Contractors:

- Big machinery (day / night / weekends)

Internal truck refill

- More visibility into suppliers’ stock - Planners will create monthly schedules shared with store men - Coordination of delivery

- Direct delivery for disruptions and planned work requires more drivers (competence and amount) and trucks (24h) - Reduction or shift of satellite employees

- All kind of tools (reduction due to centralization)

Fitters:

Contractors

Contractors:

- Big machinery (day / night / weekends)

External truck refill

- More visibility into suppliers’ stock - Planners will create monthly schedules shared with supplier/3PL

- Direct delivery for disruptions requires more drivers (change in competence not necessarily amount) and trucks (24h) - Reduction or shift of satellite employees

- Bigger and more specialized spare parts mainly for disruptions (reduction due to centralization and supplier/3PL involvement)

- Bigger and more specialized tools mainly for disruptions (reduction due to centralization and supplier/3PL involvement)

- Agreement on parameters - Automatic inventory system linked to supplier/3PL

Fitters:

Supplier / 3PL:

- Mutual visibility (demand& supply) - Coordination of delivery

Supplier / 3PL:

- Delivery of material for planned work

Supplier / 3PL:

- Spare parts for planned work

Supplier / 3PL:

- Tools for planned work

Contractors

Contractors:

(20)

20

3.3 Data Collection

The overall data collection constitutes of a combination of several quantitative and qualitative methods as well as a high amount of different data sources. In this way not only internal triangulation was achieved but also a more complete picture of the unit of analysis, the different scenarios, could be established. An overview of the data collection can be found in table 3-2 and the case study protocol is located in appendix A. In accordance with the research design process (figure 3-1) the collected data were first used to capture the research context and conduct the scenario selection. Then, data mainly served for the data analysis in order to answer the research question of this paper. An extract of the data base built base on the research design process is available in appendix E.

(21)

21 TABLE 3-2: Overview of Data Collection

Phases Methods Sources Level of detail

Asset Manager, Maintenance Team Leader North, Maintenance Team Leader

East, Maintenance Planner East, Maintenance Planner North 60 min Asset Manager, Head of Maintenance Distribution, Head of Purchasing & Logistics 60 min

Asset Manager 120 min

Head of Maintenance Distribution 60 min

Head of Purchasing & Logistics 60 min

Head of Water Supply 60 min

Focus group meeting Asset Manager, Lean Facilitator, Head of Maintenance Distribution, Head of

Purchasing & Logistics, Maintenance Team Leader East 90 min

Asset Manager I 60 min

Asset Manager II (2x) 60 min

Head of Maintenance Distribution (3x) 60 min Head of Purchasing & Logistics (3x) 60 min

Maintenance Team Leader East (2x) 60 min

Fitter North 60 min

Fitter East 60 min

Maintenance Planner North 60 min

Maintenance Engineer 60 min

Reporting and Application 60 min

Archival data

Such as flow diagrams, presentations, reports, memos, strategic policies e.g. protocol of uniform registration of failures, night call-on duty pilot overview 2013, cost reporting maintenance, order advice report

116 pages

Focus group interview 2 Fitters North, Maintenance Team Leader North ca. 120 min

Field observation 2 Fitters North ca. 120 min

Questionnaire

2 Asset Manager, Head of Maintenance Distribution, Head of Purchasing &

Logistics, Maintenance Team Leader North, Maintenance Team Leader East, Fitter North, Maintenance Planner East, Maintenance Planner North, Lean Facilitator, Maintenance Engineer

ca. 30 min

Semi-structured Interviews Asset Manager, Head of Maintenance Distribution, Head of Purchasing & Logistics,

Maintenance Team Leader North, Maintenance Team Leader East 5 x 60 min each

Solution Refinement

(purely practical) Focus group meeting

Asset Manager, Head of Maintenance Distribution, Head of Purchasing & Logistics,

Maintenance Team Leader North, Maintenance Team Leader East 60 min Explorative interviews

Framing the Problem and Solution Objectives

(research context)

Focus group meeting

Explorative interviews

Design Development (scenario selection &

(22)

22

3.4 Data Analysis

In the data analysis, completed within the design development phase, the collected data from the scenarios were reduced into categories by coding for the aim of organization and interpretation (Miles, Huberman, & Saldaña, 2014). Figure 3-4 shows the underlying coding tree of this analysis oriented on Miles et al. (2014).

FIGURE 3-4: Coding Tree

(23)

23 TABLE 3-3: Overview of Codes and assigned Resource Types

Looking into the five different scenarios separately, within-case analysis of the interaction between variables and categories was conducted to get a detailed understanding of the characteristics and patterns of each scenario (Eisenhardt, 1989). Found patterns were further compared on a cross-case basis to seek generalization of findings (Eisenhardt, 1989). Findings from the questionnaire were not coded but used to compare scenarios on a quantitative basis to provide additional insides on certain parameters. The coded qualitative data were then used to explain the shown differences and similarities.

To ensure data quality according to Yin (2009) different safeguard measures were taken throughout the data collection and analysis as shown in table 3-4:

TABLE 3-4: Ensuring Data Quality

Test Applied tactic Research step

Construct validity

- Triangulation: 30 individual and group interviews with 15 employees, questionnaire, archival data, field observation - Chain of evidence: Case study protocol, data base, research design process

Data collection (3.4) Overall (3.1, appendix A-C, E)

Internal validity

- Pattern matching within and across five scenarios

- Explanation building: Establishment of twelve mechanisms

Data analysis (3.5, 5) External

validity

- Generalizability of results: Theory Research design (3.1-3.3)

Reliability - Case study protocol - Case study database

- Records, transcripts & clarification by interviewees

Data collection (3.4, appendix A –C, E)

Descriptive code (Second order code) Interpretive code (Third order code) Main maintenance resource

Connected maintenance resource

Employee based data collection Existing data spectrum Information systems

Communication / Information exchange Disruption information

Scheduling / Planning Suppliers' transparency Stock adminitsration

Work division / specialization Responsibility of driving Work roles / skills Logistics of manpower Manpower capacity shift Contract labour Sizing of manpower Contractor costs Location of spare parts Logistics of spare parts Sourcing procedures

Sourcing focus Quality of spare parts Manpower, Information Sourcing based stock reduction

Amount of spare parts locations

Location of tools Logistics of tools Contract tools

Amount of tool locations Quantity / Cost of tools Manpower Quantity / Cost of spare parts

Availability of tools

Quantity / Cost of manpower

Manpower Flexibility of manpower

Productivity of manpower

Spare parts

Manpower, Spare parts Manpower

Manpower Spare parts

Availability of spare parts

Availability of information

Information support of manpower

Coordination of manpower

Management of spare parts

(24)

24

4 FINDINGS

Based on the data collection and analysis several findings could be detected that hold significant value for answering the underlying research question. The following four sections focus on the findings of this research organised by the four types of maintenance resources: Information, manpower, spare parts and tools. Each of these sections will pay attention to the different mechanisms found in the context of the respective resource type. By doing so, the five different scenarios of the supply network will be compared, the mutual impact among mechanisms exposed and the relevance of each mechanism for the three different SCRES phases uncovered. A final section will summarize all findings across resource types and scenarios.

4.1 Findings regarding Information

Starting with a more quantitative view on information, figure 4-1, located at the end of this section, contrasts the four scenarios regarding their envisioned increase of availability and accuracy of information compared to the as is situation. All four scenarios are expected to improve in this aspect, with the scenario of CW pick-up and small cabins showing the highest expected increase. To offer an explanation for these aspects, the qualitative findings are further consulted. Here, first, it has been shown that most data that have been selected for the data analysis could be assigned to the resource type of information. This shows the significant relevance of this resource type and results in the highest amount of mechanisms (table 4-3). Availability of information, information support of manpower, coordination of manpower and management of spare parts have been detected in this context.

The mechanism availability of information established in the data analysis deals with both, the existing data spectrum, i.e. the respective information the focal company decided to consider generally and the employee based data collection (aided by software and hardware) for these information. Looking into the scenario comparison in table 4-1 in the section on overall findings, a clear difference between the supply network of the as is situation and the other four scenarios can be seen. In the base scenario the availability of information is hindered by the limited existing data spectrum: “The costs of used resources are not separate for single

disruptions” (Maintenance Engineer) and the poor employee based data collection: “But I also think if you expect fitters to do too much, which is not actually their job it is too much administration at some point. And if they don't do it, the whole thing is not valuable anymore”

(Fitter East). Improved availability of information can be found in all other scenarios due to a wider data spectrum and technology aided data collection: “The fitters could use their iPad to

collect more disruption specific data” (Maintenance Engineer). Considering the dependencies

(25)

25 are dependent on the availability of information. Referring to table 4-3, it is noticeable that the availability of information is equally important in the preparation, response and recovery phase. This can be explained by the need for spare parts administration (data collection) in the preparation, the demand for transparency (existing data spectrum) in the response and the need for disruption reporting (data collection) in the recovery.

Information support of manpower is another mechanism found from the data analysis and deals with information systems and communication. From the scenario comparison in table 4-1, it can be seen that information systems are equally well available throughout all scenarios. They enable a faster response by giving immediately access to information like location of pipes or automatic calculations. “Without the iPad I can do nothing“ (Fitter East). Communication somewhat differs between the scenarios as different designs result in an increased or decreased need for contact between parties. Although the knowledge will be centralized at the new central warehouse in all future scenarios, only with the CW pick-up the existing close communication will be kept. “Negative is that there is less connection with the

planners” (Head of Purchasing & Logistics). Nevertheless, the more decentralized position of

fitters in these scenarios enables exchange with colleagues. “You need to talk to the

colleagues because now at the job you must know what is happening. You need to see each other and I think places like that [small cabins] would be fine for that” (Fitter East). From table

4-2 it can be seen that the productivity and flexibility of manpower is dependent on the information support. Referring to table 4-3, it is noticeable that the information support of manpower is mostly represented in the preparation (establishment of knowledge) and response (disruption oriented communication and aid from information systems).

The third detected mechanism deals with the coordination of manpower. It is linked to disruption information and scheduling. Table 4-1 reveals that in all five scenarios disruption information is satisfactory. “When the information is going to the planner that the information

is so clear that he can put it directly to the fitter” (Maintenance Team Leader East). Scheduling,

however, is believed to improve in all four future scenarios due to centralization (one team). “But I think especially for the planning, it is this and all situations when they sit together, they

can make a better planning and they can make the plan for all the plumbers together. And I think that is the big win” (Maintenance Team Leader North). Coordination of manpower

(26)

26 management of spare parts due to low suppliers' transparency, poor stock administration and short term schedules. Parameters for stock administration as well as information about withdrawal are available at the central warehouse but neither at the satellites nor at the trucks. This is on the one hand linked to the absence of technology for stock administration, but one the other hand simply by a lack of standards. “Fitters take material, but you don’t know what is

in their cars. Fitters’ cars are black boxes, there is no visibility into this” (Head of Purchasing &

Logistics). In contrast the four future scenarios show an improved management of spare parts due to increased suppliers' transparency, improved stock administration and long term schedules. “Like the planning also, they must look further ahead for us to see what material

we need much earlier so they can get it ready in the warehouse.” (Fitter East). FromTable 4-2 it can be seen that the management of spare parts impacts their availability, quality as well as quantity. Further, table 4-3 points out that this mechanism is mainly represented in the preparation as it is the phase were spare parts need to be planned.

4.2 Findings regarding Manpower

Beginning with the quantitative findings again, figure 4-1 shows the expected reduction of manpower cost for all five scenarios. Here, although an increase in efficiency for all future scenrios can be seen, the differences between all scenarios are rather insignificant. Offering again an explanation based on the qualitative findings, it can be seen that also manpower is a resource that holds a great relevance as it is the resource type with the second most data and mechanisms (table 4-3). For manpower the mechanisms flexiblity, productivity and quantity respectively costs have been found to influence SCRES.

Flexibility of manpower is the first of the three mechanisms found for manpower. It deals with work roles respectively skills, capacity shift, logistics and contract labour. Table 4-1 shows that minor differences exist between the base scenario, the CW pick-up and the three remaining scenarios. In the base scenario high flexibility is given due to individual logistics (1 truck per fitter), flexible workload, cooperation with contractors and low specialization, but also hindered by a lack of skills. “Now they are planned, but they are doing work, when you call them they

can come” (Head of Maintenance Distribution). “And that is a complaint that I heard earlier about from fitters to the men who work at the warehouse, so they have no feeling with the working fields.” (Maintenance Team Leader East). The CW pick-up supply network is most

similar to the base scenario in terms of flexibility. Here only a slightly lower flexibility is given due to minor changes regarding specialization. For the remaining three scenarios a lower flexibility is assumed due to higher specialization. “And that is really difficult to give me a water

(27)

27 productivity and has further a mutual dependency with the quantitity of available manpower (table 4-2). Flexibility is also dependent on the availability of spare parts and tools. In the preparation phase work roles and skills are set and shiftable capacity for disruptions considered, whereas in the response contractors and logistics play a major role (table 4-3). Productivity was already mentioned as another mechanism of manpower. It deals with work division respectively specialization in general and with the responsibility of driving in particular. As productivity can be seen as the counterpart of flexibility all previously mentioned aspects are reflected. Considering table 4-1, the base scenario shows low productivity due to low specialization, in particular regarding driving. “So you need the men. So I look for some other

work [...] that is not a core, these are jobs to refill the gaps” (Head of Purchasing & Logistics).

“The costs and the time. When I have to sit a lot in the car, I cannot do work” (Fitter North). For the CW pick-up supply network only a slightly higher productivity due to minor changes regarding specialization and driving can be detected. For the remaining three scenarios a higher productivity due to higher specialization, in particular regarding driving, is given. “They

bring it to me on the job. That would be the best for me because then I can focus on my job and not on the materials and not on the driving. Sometimes it takes you about one hour so you lost one hour” (Fitter East). As already mentioned productivity stands in contradiction with

productivity and has further a mutual dependency with the quantity of available manpower (table 4-2). Moreover, productivity is also dependent on the availability of spare parts and tools. Lastly, quality of spare parts plays a role regarding the installation time. For the same reasons as flexibility, productivity is most relevant in the preparation and response (table 4-3).

The last discovered mechanism for manpower is quantity respectively cost of manpower. Here internal sizing as well as contractor costs matter. Table 4-1 shows that only minor differences exist between the scenarios. In the base scenario the quantity of manpower is given by decentralized teams, existing demand and use of contractors. “You cannot oversee

the costs because, maybe there is not a big problem but it is here in the town under the road or you have to saw the asphalt, other contractors have to come to help” (Maintenance Team

Leader North). In the CW pick-up and the external truck refill supply network the internal amount of fitters and storemen decreases due to centralizations, but at the same time a small increase for drivers is necessary. For the external truck refill additional costs are given. “I think

you have more external costs here, because you hire a firm who provides you of material and equipment” (Maintenance Team Leader North). For the small cabins and the internal truck refill

also the quantity decreases due to centralizations, but because also logistics activities are centralized more drivers are needed. “So then you have to organize here, the people who will

bring it” (Maintenance Team Leader East). The dependencies of quantity of manpower with

(28)

28

4.3 Findings regarding Spare parts

Figure 4-1 indicates the expected reduction of the costs of spare parts for all five scenarios. An increase in efficiency for all future scenarios compared to the as is situation can be found. Further, there are significant differences between all future scenarios. The scenario of CW pick-up is showing the highest reduction. These differences can be explained with help of the qualitative findings. Three mechanisms, namely availability, quality and quantity respectively costs of spare parts have been identified as linkages between spare parts and SCRES. The mechanism availability of spare parts found in the data analysis regards mainly the location of spare parts, but also their logistics as well as sourcing. Table 4-1 shows that the base scenario as well the external truck refill are different to the other three scenarios. In the base scenario the availability of spare parts is mainly assured by the higher amount of locations, but hindered by poor logistics. Different kinds of spare parts are available either in the central warehouse, satellites or fitters’ trucks. “The plumbers they always have for certain disruptions

they have the tools and also the materials in the van and I think 80% of all the things they can repair” (Maintenance Team Leader North). For the supply network scenarios of small cabins,

CW pick-up and internal truck refill the availability is lower due to reduced locations, but counteracted by improved logistics. For the response to a disruption spare parts could be quicker available when always delivered by the central warehouse. “For the bigger

interruptions I think the possibility to bring directly from the warehouse. So the fitters can do their work. It will reduce the costs of of the interruption” (Maintenance Planner East). For the

external truck refill the dependency on the external party further reduces the availability. Table 4-2 shows the already mentioned dependencies and adds the mutual dependency with quantity of spare parts. In the preparation phase the location of held stock plays the main role, whereas the logistics of spare parts mostly dominate the response phase as material has to be moved to disruption sites (table 4-3).

Quality of spare parts is a found mechanism for the preparation (table 4-3). High quality of spare parts is given in all scenarios due to the exiting sourcing focus (table 4-1). Quality issues can extend the time it takes to install a spare part in the repair and have thus influence on the response. “And sometimes spare parts are chosen on the money, because sometimes it is

push fit. You put it in and it is ready, but sometimes you have to pull it out of each other, get a ring above it and then you have to clean it and then you can pull it in. it is more complex. So then you can sometimes win time and costs when it is easier” (Maintenance Team Leader

East). Due to its connection with fitters’ repair, quality of spare parts is connected with productivity of manpower as already mentioned (table 4-2). Further, there exist a mutual dependency with the quantity as high quality material is expected to last longer.

(29)

29 scenario offers the highest quantity due to the most locations. The scenarios of small cabins, CW pick-up and internal truck refill are characterized by a decreased quantity due to centralization and new sourcing startegy (e.g. modularity of material). “Different parts that can

be combined for different problems” (Head of Purchasing & Logistics). “With the new central warehouse we need less inventory” (Head of Water Supply). The scenario of external truck

refill shows a highly decreased quantity as here not only centralization, new sourcing strategies but also the involvement of the supplier respectively 3PL plays a major role. All dependencies with other mechanisms were already mentioned and can be seen in table table 4-2.

4.4 Findings regarding Tools

Viewing tools first from an quantitative angle, figure 4-1 depicts that different to all other resources, the reduction of tool cost in the scenario of small cabins is actually lower than in the base scenario. CW pick-up is also for this resource the most efficient scenario. This distribution can be explained with the qualitative findings. For tools the mechanisms availability and quantity respectively costs have been detected.

The mechanism availability of tools established in the data analysis deals mainly with the location of tools, but also their logistics as well as contract tools, i.e. tools provided by external parties. Looking into the scenario comparison in table 4-1, clear differences between the supply network of the as is situation and the other four scenarios can be seen. In the base scenario the availability of tools is assured by a high number of locations and by contractors. “The basic

tools are in the truck, but there are also some specials and they are located in Winschoten or here in Groningen or in the central warehouse” (Maintenance Team Leader East). In the

scenario of the small cabins a higher tool availability is given due to decentralized locations. “The bigger machines, the pumps and that kind of stuff, they can put in here. That is the big

advantage” (Head of Maintenance Distribution). For the other scenarios a lower availability of

tools is the result of only one tool location apart from the fitters’ trucks. Availability of tools holds a mutual dependency with the quantity (costs) of tools as one conditioned the other (table 4-2). Referring to table 4-3, it is noticeable that the availability of tools is mainly important in the preparation phase, as here tools have to be provided to be usable in case of a disruption. Lastly, quantity respectively costs of tools is directly linked to tool locations in the preparation which were already discussed in the previous paragraph. Thus, quantity of tools decreases with centralization and increases with decentralization: “If you going to the central warehouse

I think some tools you can reduce” (Head of Purchasing & Logistics). For the dependency to

(30)

30

4.5 Overall Findings

(31)

31 FIGURE 4-1: Scenario Comparison of Quantitative Findings: Efficiency & Redundancy2

FIGURE 4-2:Scenario Comparison of Quantitative Findings: SCRES Phases

2_{Costs of outsourcing a part of the logistic activities in the external truck refill scenario are not considered. The numbers only}

(32)

(33)

33 TABLE 4-2: Mechanisms, their Dependencies and Linkage to the Resource Types

Availability of information Information support of manpower Coordination of manpower Management of spare parts Flexibility of manpower Productivity of manpower Quantity / Cost of manpower Availability of spare parts Quality of spare parts Quantity / Cost of spare parts Availability of tools Quantity / Cost of tools Availability of information Dependancy on avalable information Dependancy on avalable information Dependancy on avalable information Dependancy on avalable information Dependancy on avalable information Dependancy on avalable information Dependancy on avalable information Dependancy on avalable information Dependancy on avalable information Information support of manpower Dependancy on information Dependancy on information Coordination of manpower Dependancy on coordination Dependancy on coordination Dependancy on coordination Management of spare parts Dependancy on management Dependancy on management Dependancy on management Flexibility of manpower Contradiction Mutual dependancy Dependancy on available spare parts Dependancy on available tools Productivity of manpower Mutual dependancy Dependancy on available spare parts Quality influences working time Dependancy on available tools Quantity / Cost of manpower Availability of spare parts Mutual dependancy

Quality of spare parts Mutual

dependancy

Quantity / Cost of spare parts

Availability of tools Mutual

dependancy

Quantity / Cost of tools

Tools Resource types / mechanisms

Information

Manpower

Spare parts

Tools

(34)

34 TABLE 4-3: Efficiency and Redundancy of Mechanisms in the SCRES Phases

Efficiency Redundancy Total Efficiency Redundancy Total Efficiency Redundancy Total

Information 12 82 93 4 57 61 3 21 25 179

Availability of information 3 10 13 4 9 13 3 11 14 40

Information support of manpower 1 23 24 20 20 4 4 48

Management of spare parts 8 31 18 2 26 6 41

Coordination of manpower 18 39 26 2 6 50

Manpower 31 58 89 16 29 45 2 3 5 139

Flexibility of manpower 10 25 35 2 21 23 3 3 61

Productivity of manpower 16 12 28 14 6 20 2 2 50

Quantity / Cost of manpower 5 21 26 2 2 28

Spare parts 23 46 70 8 16 23 3 3 96

Availability of spare parts 13 43 56 8 15 23 3 3 82

Quality of spare parts 3 3 1 1 4

Quantity / Cost of spare parts 10 10 10

Tools 7 17 24 8 8 2 1 3 35

Availability of tools 2 17 19 8 8 2 1 3 30

Quantity / Cost of tools 5 5 5

Total 73 203 276 28 110 138 7 28 35 449

Total Mechanisms / SCRES Phases /

Efficiency & Redundancy

(35)

35

5 DISCUSSION

It is known from literature that maintenance can be a facilitator of SCRES (Saleh & Marais, 2006) and thus, of an efficient as well as effective response to unforeseeable disruptive events in the supply network (Christopher & Peck, 2004; Ponomarov & Holcomb, 2009; Scholten et al., 2014). What is not clear to date is how exactly to satisfy both the efficiency and effectiveness aspect within the link of both concepts. Redundancies of resources, as in maintenance, that help to deal with disruptions are seen as an enabler of SCRES (Sheffi & Rice Jr., 2005), however, supply networks including maintenance, are often designed to optimize cost and efficiency (Marais & Saleh, 2009; Marais, 2013; Rosqvist et al., 2009). Although also maintenance resources are costly, strategies based around lowest cost reduce resource buffers and thus, might impact the networks vulnerability and hence SCRES (Blackhurst et al., 2011; Ivanov et al., 2014). So the strategic fit between efficiency and effectiveness of SCRES can be achieved by actively designing the balance of maintenance resources regarding efficiency and redundancy (Christopher & Peck, 2004; Scholten & Schilder, 2015) into a supply network (Bakshi & Kleindorfer, 2009; Blackhurst et al., 2011; Ponomarov & Holcomb, 2009; Scholten et al., 2014). To answer how to design resources in maintenance to balance efficiency and redundancy for supply network resilience, the proposed conceptual framework was evaluated in the context of the supply network of a single company. SCRES, as defined for this paper, consisting of its three phases from the disruption profile as defined by Ponomarov & Holcomb (2009), Sheffi & Rice Jr., (2005) and Tukamuhabwa et al. (2015), was combined with the concept of maintenance resource structure, gathered from Balzer & Schorn (2015), Hastings (2015) and Kelly (2006) amongst others, which includes four different maintenance resource types and their structuring decisions.

(36)

36 FIGURE 5-1: SCRES Decision Making Model

Information Manpower Spare parts Tools Information Manpower Spare parts Tools Information Manpower Spare parts Tools Availability of information Information support of manpower Coordination of manpower Management of spare parts Flexiblity of manpower Productivity of manpower Quantity / Cost of manpower Availability of spare parts Quality of spare parts Quantity / Cost of spare parts Availability of tools Quantity / Cost of tools

Efficiency: Small increase Efficiency: Big increase Redundancy: Small increase Redundancy: Big increase

(37)

37 5.1 Resource Balancing in the Preparation