Maintenance of the water distribution network of Waterbedrijf Groningen

Maintenance of the water distribution network of

Waterbedrijf Groningen

How to improve the efficiency and effectiveness of the maintenance process?

Faculteit Economie en Bedrijfskunde

Master Thesis

Author: Alex de Groot Boersma

Maintenance of the water distribution network of

Waterbedrijf Groningen

How to improve the efficiency and effectiveness of the maintenance process?

Master thesis

Technology Management

Company: Waterbedrijf Groningen University: Rijksuniversiteit Groningen

Study: Msc. Technology Management Supervisor: Drs. ing. H.L. Faber

Author: Alex de Groot Boersma Second supervisor: Dr. ir. W. Klingenberg

Student number: s1344242 Company supervisors: Dhr. J. Breet

Dhr. H. Zwartenkot

Date: 16/03/2011

Preface

This thesis is to finalize my study Master of Science Technology Management at the Rijksuniversiteit Groningen. I got the opportunity to conduct research at Waterbedrijf Groningen through an internship. This research was focussed on the maintenance process of the water distribution network in the province of Groningen. I want to thank the people who contributed to this report and helped me to complete this project.

There are many people who have supported me throughout this research, and I would like to start by thanking my supervisors at Waterbedrijf Groningen. Hans Breet, Henk Zwartenkot and Henry Korteschiel who have taken the time to give me feedback and gave me insight in the organization. Also thanks to the employees at the distribution department for sharing their knowledge and expertise with me.

Also, I would like to thank my supervisors at the university, Mr. Faber for giving me feedback and understanding for my situation during my research and Mr. Klingenberg for his second supervisory work. Last but not least, I would like to thank my parents, friends and girlfriend for supporting me and their good advices.

Management summary

The research described in this thesis is performed at the Distribution department (OHD) of Waterbedrijf Groningen. The company produces and distributes drinking and industrial water throughout the province of Groningen. The Distribution department is responsible for all maintenance of the distribution assets, starting at water distribution pipelines after the production locations up to the water meters at the customers. Partly as a result from previous study at the department, that was carried out by other students of Technology Management, this research has started. Outcomes are, to some extent, related to the outcomes of this research. However, this research is more extensive, new insights are gained and more diverse actions are proposed. Next to this, Waterbedrijf Groningen wants to stay in the top 3 of the benchmark carried out by VEWIN, a governmental institute to compare costs and knowledge between water companies about water and its business processes. Since these two aspects influenced the vision of the department, the management wanted to professionalize. Hence, this research was started to define and select the most important bottlenecks that influences the performance of the maintenance activities. All the possible aspects that influence the efficiency and effectiveness of the maintenance work are identified. Thereafter, solutions are given to improve the situation. The research question that ensued from this is defined as following:

What actions need to be taken to improve the efficiency and effectiveness of the maintenance activities performed at the Distribution department of Waterbedrijf Groningen?

To gain overall insight in the company, this thesis starts with a general description of the company, the Distribution department and its activities. After defining the problem statement, data is gathered about the variables that influence the efficiency & effectiveness of the maintenance activities. Interviews are taken to orientate on problem areas. More different interviews, data collection and observations made the problems more obvious. To gain insight in efficiency and effectiveness there is a need for

indicators to measure the performance of the Distribution department. Tracking the performance of

The most important findings are summarized below where especially of interest will be the field of Inventory Control (IC). This aspect has a substantial impact on the improvement of the situation. Inventory Control:

As mentioned above, the first factor that influence the performance of the maintenance activities is referred to the way inventories are managed. The diagnosis has revealed that the inventory control systems are inefficiently managed. For this reason, most of the priority of this research is dedicated to this aspect. A factor that determines inventory efficiency are Stock Turns. This factor explains that the inventory level of the district inventories is very high. The Stock Turns are quite low, which is an indicator of the high level of inventory, because it means that material is quite long in stock.

For the maintenance mechanics it is essential to always have spare parts available when failures occur. Because of inaccurate registration of used material, the inventory levels are often incorrect. Direct consequence is that material is out of stock which leads to problems to execute the work. In this research it is concluded that the ordering process is inaccurate because of incorrect information of inventory levels. Hence, first the administration of inventory levels needs to be correct. The administration can be improved by introducing scanning equipment to register the input and output of the warehouses. This will also reduce the chance of failures in the registration process. Accountability of the mechanics, who will work with the system, is required.

The next step has to be the focus on the method used for ordering the material in order to increase the material service level and reduce the investment on inventory. Ordering models will enhance this. An appropriate model that is suitable for the situation of Waterbedrijf Groningen is (Q,r) model, where the inventory is continually reviewed. When inventory levels will come below a certain level, the material will be reordered with the most feasible order quantity. After evaluating different methods to implement this model, it is recommended to calculate the reorder moments and order quantities with queries. These are written spread heads in excel and have to be filled manually.

Motivation

situation. Moreover, it is stated by management that there is an “islands culture” where the total view of value driven processes are lost. Everyone is busy with their own tasks which can result in mutual adjustments that are inadequate. This can cause inefficient execution of work, higher chance of failures and more repairs in the distribution system. Last but not least, resistance to change need to be overcome. Change interventions can fit into one or more strategies: communication & education, participation & involvement and facilitation & support.

Planning and scheduling

The last factor to improve the situation is planning and scheduling of the maintenance work. More standardization in planning will increase the effectiveness of the planning. Currently, the planning is based on experience and it is proved by best-practice research of the researcher that in maintenance environments that planning with standard times enhances the capacity planning and day planning. By planning with standard times it is estimated that productivity can be increased with 10 percent as perceived from the benchmark study, literature and interviews with planner. The data to calculate these standard times needs to be gathered by observations. Next to this, a digital work order system is recommended to improve the transparency and registration of work. Transparency is needed to see for everyone which work is performed when and by whom. The registration is needed as an input for Asset Management that makes strategic decisions for the long term maintenance planning.

Conclusion

Definitions

Benchmarking

The continuous, systematic search for, and implementation of,

better practice that leads to better performance.

Brainstorm

A basic problem solving tool where the creativity of a group

to open up new possibilities. It uses the unevaluated ideas of

groups to generate list of: problems, causes, solutions and

actions.

BSC

Balanced Scorecards. A framework by which a strategy can

be translated into operational terms, where measures of past

performance are enhanced with measures of the drivers of

future performance.

CMMS

A computerized system to assist with the effective and

efficient management of maintenance activities through the

application of computer technology.

Condition based maintenance The preventive maintenance initiated as a result of knowledge

of the condition of an item derived from periodic, routine or

continuous monitoring

Control

Control is the process by which comparisons are made

between the plan and the performance, either during or after

execution.

KWR

Water cycle Research Institute that performs research on risk

analysis and effects and is commissioned by VEWIN

KPI

Key Performance Indicators. A measure of performance. KPIs

are used to evaluate the success of a particular activity in

which it is engaged

Maintenance

The combination of all technical and associated administrative

actions intended to retain an item, or restore it to a state in

which it can perform its required function.

Modification

Replacement of distribution assets.

NBD

Engineering department for new distribution projects.

OBIS

Digital maintenance control system.

Process

A collection of timely ordered elements, most often called

events or activities.

OHD

Maintenance department of the distribution network.

OLM

Underutilized delivery minutes (Ondermaatse leverings-

minuten) The average number of minutes per year that a

customer has a water pressure below 50 kPa.

TKC

Customer contact department (Technische klantencontacten),

maintains the customer contacts for incoming complaints.

VEWIN

Governmental institute to compare costs and knowledge

between water companies about water and business processes.

WBGR

Waterbedrijf Groningen

Table of content

Preface ... 3

Management summary ... 4

Definitions ... 7

1.

Introduction ... 10

1.1

Background of the research ... 10

1.2

Research outline ... 11

1.3

Foundation of the company ... 12

1.4

Company mission statement ... 13

1.5

From “source to sink” ... 13

1.6

Maintenance of the distribution network... 14

1.7

Cost price of drinking water ... 15

2.

Problem Statement ... 17

2.1

Research question ... 17

2.2

Definition efficiency and effectiveness ... 17

2.3

Methodology ... 18

2.4

Sub questions ... 20

2.5

Conceptual model ... 20

2.5.1

Input factors ... 22

2.5.2

Supporting factors ... 24

3.

Diagnosis ... 27

3.1

Maintenance process ... 27

3.2

Maintenance of distribution assets ... 30

3.3

Measuring performance of the maintenance work ... 32

3.4

Maintenance regimes ... 35

3.5

Quality material and equipment ... 36

3.6

Balance of in-house capability and outsourced service... 37

3.7

Planning and problem solving ... 37

3.8

Material flow and inventory control ... 40

3.8.1

Inventory process ... 40

3.8.2

Aspects influencing inventory efficiency ... 42

3.9

Organizational climate ... 44

3.9.1

Culture and motivation ... 45

3.9.2

Culture and cooperation ... 46

3.9.3

Willingness to change ... 46

3.9.4

Capabilities of employees ... 46

3.10

Conclusion of the diagnosis ... 48

4.

Design ... 50

4.1

Problem statement design phase ... 50

4.2

Inventory control system ... 52

4.2.1

Administration of inventory ... 52

4.3

Motivation and behaviour of employees. ... 56

4.3.1

Motivation ... 57

4.3.2

Skills ... 60

4.3.3

Willingness to change ... 60

4.3.4

Conclusion motivation and behaviour of employees ... 62

4.4

Planning and scheduling ... 63

4.4.1

Standard job times ... 64

4.4.2

Driving time reduction ... 66

4.4.3

Work order system used for planning ... 67

4.4.4

Digital planning system ... 69

4.4.5

Conclusion planning and scheduling ... 69

5.

Conclusions and recommendations ... 70

6.

Reflection and Further research. ... 73

References ... 74

Appendix A.

Organization chart Waterbedrijf Groningen ... 77

Appendix B.

Benchmark study at other companies... 78

Appendix C.

Process Flows maintenance planning ... 80

1. Introduction

Water is important for everyone. We use water on a daily basis to drink, take a shower, flush the toilet or use a washing machine. It‟s also important for the industry. There is a growing need for industrial water and water purification. Waterbedrijf Groningen delivers high quality and reliable drinking water in the province of Groningen and a small part of Drenthe. The joint venture of Evides Industrial Water, Waterbedrijf Groningen and Drenthe is called: „North Water„ and provides solutions for industries on special treatment of drinking water, industrial water and waste water. The drinking water process starts with the production of water at 6 production locations that filters water from surface water and ground water. The water is distributed throughout a pipeline network with a total length of about 5000 km. Thereafter, the water comes through 4000 km of connection pipes connected to the residential buildings and companies. The last asset in this process is the water meter at which water consumption of the customer will be measured.

1.1

Background of the research

This master thesis is written to finalize my study Technology Management at Rijksuniversiteit Groningen. The purpose of this paper is to analyze the organization around the maintenance of all the assets in the water distribution network throughout the province of Groningen. The network starts at water distribution pipelines after the production locations up to the water meters at the end consumers. The company is a non-profit organization that wants to keep their costs as low as possible to maintain the lowest water prices of the Netherlands. Thereby the service to the customer is of equal importance. The goal of WBGR is to stay in the top three of the benchmark carried out by VEWIN, a governmental institute to compare costs and knowledge between water companies about water and its business processes. The costs of maintenance and renewal of the network is rising because of the decreasing condition of the network. For this reason, the department of maintenance and distribution (OHD) wants to know in what way their activities can be more efficient and effective.

There are different signals of the department that indicates a negative influence on efficiency and effectiveness of the department. Previous research that is carried out by four students of Technology Management (January 2010) indicates four factors that need to be considered for improvement of the planning of maintenance activities. These are: available material, standard times, systems and

quality of input information for planning. These outcomes are taken along in this research. The

that it is a process of continual improvement where an organization is constantly reviewing and revising the system. However, most of the work that is done is based on „ad-hoc decisions‟ and by living in the illusion of the day. There is little insight in the status of maintenance and when there are no complaints of the customer the work is performed well. Also, in order to comply with the ISO 9001 norms the maintenance is currently not sufficiently in control.

1.2

Research outline

The methodology used in this thesis is based on the DDC-model (De Leeuw, 1996). This means:

Diagnosis of the problem situation, Design for solution and Changing the situation. This

methodology is based on logic reasoning in order to improve problem situations. Roughly, this thesis is composed of five parts as visually represented the figure underneath where the DDC phases are included. Chapters: Research overview D ia gn os is D ia gn os is R ec om m en da tio ns R ec om m en da tio ns D es ig n D es ig n P ro bl em st at em en t P ro bl em st at em en t In tro du ct io n In tro du ct io n Current situation Waterbedrijf Gronigen Initial problem statement Introduction to the problem Research design and planning Conceptualization of the problem Description of the scope of the research Objective judgement Main research objectives Define direction and selection criteria Design Generate alternatives Priorization of research objectives Conclusions and Recommendations Determine solutions Chapter 1 Chapter 2 Chapter 3 Chapter 4 Chapter 5

Fig 1.Research overview project

price of drinking water. Asset management is closely related to the maintenance department because the work of maintenance is input for the decisions of asset management and these decisions leads to a better condition of the network and thus less work for the maintenance department. Chapter 2 describes the problem statement where the methodology of this research is explained and the planning of this project is motivated. Next, the problem is formulated in a research question and sub questions. A conceptual model is created to present the factors that influence the main problem. These factors are explained afterwards. In the next chapter the diagnosis is carried out according to the sub questions of the conceptual model. This chapter describes the current situation where the factors in the conceptual model are included. Thereby, the problems that emerged during the interviews and observations are explained. This diagnosis is also supported with scientific literature and books that are referenced in this thesis. The last part is design of the problem situation which describes how to solve or improve the situation. Alternatives are analyzed and best options are concluded. Afterwards, these statements are concluded and recommendations are given.

1.3

Foundation of the company

The history of the WBGR begins with the need for cleaner drinking water. In the 19th century people drank their water from wells and canals. People made their own water supplies with rain barrels or larger communities gained their water from governmental initiatives who made wells and pumps. At that time, there was nobody who had influence on the quality of the water and nobody wondered themselves if this was healthy drinking water. Until 1888, when a large epidemic outbreak of Cholera and Tyfus came across the Netherlands. Large numbers of people died except for the people in Amsterdam because they had a central source of drinking water distributed into the city. Research indicated that the reason of this epidemic outbreak was the quality of drinking water. Thereafter a number of initiatives started to set up a supply of clean drinking water.

there are two district offices from where the maintenance of the distribution network is executed. There are five organizational sectors that are responsible for operating the organization in an appropriate way. The organisation chart is shown in appendix A.

1.4

Company mission statement

The following mission statement is formulated by the company:

Waterbedrijf Groningen wants, as a social organization, to sustain the water interest in the province of Groningen for the present and the future.

The company is not only a supplier of different kinds of water. They also contribute to the public health, conservation of nature and environment, creating sustainable sources, employment and innovation. The strategy to reach this is cooperation with other partners in water sector, knowledge institutes and environmental organizations. The goal is to create synergy. This can be for delivering high quality water at a reasonable price, protection of nature or creating a good business climate for regional organizations. The water company is a social responsible company that wants to create value for the region in the future by continually improving their products and services.

1.5

From “source to sink”

Every food and drinking industry is subjected to stringent regulations in terms of safety and hygiene. Also for drinking water there are pre-described regulations and specifications. There is a special law called: ”drinking water legislation” that describes the maximum amount of elements and the prohibited elements in the water. During the production of the water these quality norms are controlled by the laboratory called „Waterlaboratorium Noord (WLN)‟ which analyzes the quality of the drinking water for WBGR and „Waterleidingmaatschappij Drenthe‟.

Fig 2. Water distribution flow (Intranet WBGR)

After the production locations, there are six distribution pumping stations and two water towers that serve as extra buffer and pressure. In this way, the whole supply area will be reached with enough pressure and quantity.

1.6

Maintenance of the distribution network

This research is initiated by the maintenance department (OHD). Therefore, this section describes the activities of this department. The maintenance department is responsible for the condition of the main network, connection network and all the objects within this network. The core business is to deliver an continuous water supply with enough pressure and quantity using the required quality of water. As described in the benchmark, the distribution process consists of: maintenance of main pipes, maintenance of connection pipes and replacement of water meters. This department is a part of the „sector water supply‟ and contains 51 FTE. The department is subdivided into three sub-departments, team north, team east and special services. Team north and east provides maintenance in respectively the north-west and the south-east part of the province. Special services provide support to both maintenance teams. Thereby the department is responsible for a reduced quality of the water caused by pollution in the network. This can be caused by corrosion or bio fouling of the water pipes or by modifications of maintenance mechanics. The department has a 24 hour service in order to come into action when a breakdown occurs any time of the week. This will guarantee the continuous delivery of reliable drinking water with enough pressure and quantity.

monitors and valves are done every 4 year. The preventive maintenance also entails KLIC reports and fixed inspections of different objects in the network. This can be either cyclic based inspections or condition based inspections. The department provides preventive maintenance activities that involves the KLIC reports. These are reports that have to be made when someone has to excavate with machines. These reports clarify where and what kind of networks are located in the ground and therefore prevent breakdowns. Next to this, the department carries out attendance work when there are risks for water tubes at excavation work. This is done by the team special services. For the KLIC reporting system there are more than 15.000 KLIC reports processed every year. Thereby, every year 1.500 KB (Cathodic Protections) locations are inspected and repaired if needed.

Curative is based on the maintenance carried out after a failure has occurred and intended to restore an item to a state in which it can perform its required function (Kister, Hawkins, 2006). When breakdowns in the network occur, the effect can be a termination of water supply to the customers. The maintenance mechanics have to solve these problems. These actions are not foreseen and therefore cannot be planned and scheduled. On the basis of severity of the situation, the urgency of the work will be determined. There are 300 distribution pipe failures on a yearly basis. This is relatively low compared to other companies, largely due to the type of soil in the province of Groningen.

1.7

Cost price of drinking water

Up until now the price of drinking water of WBGR is the lowest in the Netherlands. WBGR wants to keep this price low but investment costs will increase drastically because of renovation of existing production facilities and sanitation of pipes in upcoming years. To gain more transparency in the Dutch drinking water cost prices a central benchmark is created. This Benchmark compares the performance of the Dutch drinking water companies in terms of their core results. The performance is tested according to four pillars. These are: drinking water quality, service, environment and finance & efficiency. It gives shareholders, central government, customers, researchers and the companies themselves a better understanding of the performance of the drinking water companies. These stakeholders all want to have more transparency of company information and efficiency (VEWIN Andersen Consulting, 1999).

(shame). Everyone wants to prevent bad publicity. The study aims at offering water companies instruments to help them further improve their business processes. Companies have adopted each other‟s best practices over the past ten years. Since the introduction of benchmarking in 1997 water quality has continued to improve, good service has been maintained, while costs in real terms have declined. (VEWIN Accenture, 2006). That the Benchmark is bearing fruit is confirmed in research by Erasmus University Rotterdam. This shows that the sector has attained an efficiency improvement of 23% since the introduction of the Vewin Benchmark. In order to compare the costs between companies, the benchmark tries to use a universal composition of the cost price that is applied for all the Dutch water companies. The benchmark consists of operational costs, depreciation, cost of capital and taxations. The operational costs involve the Distribution process costs. Therefore it‟s interesting to look at these costs. The costs are divided into six main processes incorporating the maintenance costs. These costs include maintenance of distribution pipes, connection pipes and change and maintain water meters. The figures of the benchmark of 2009 show lowest costs in the distribution process of WBGR. This is due to lower depreciations of assets and cost of capital compared to other companies. A second reason is the small number of failures in the network compared to other companies. The operational costs of the distribution process are therefore on average.

2. Problem Statement

In this section, the problem statement is formulated in terms of a main research question. Thereafter the methodology will be described involving the necessary steps to answer the main research question. Next, the sub questions are determined by means of interviews and personal interpretations of the problem, while matching the needs of the management of WBGR. The conceptual model is presented and discussed in order to identify what needs to be diagnosed subsequently.

2.1

Research question

At Waterbedrijf Groningen there is a need to improve the efficiency and effectiveness of their maintenance process. Eventually, this has the purpose to reduce operational spending. The Main Research Question is formulated in cooperation with the management of the OHD department and is defined as following:

What actions need to be taken to improve the efficiency and effectiveness of the maintenance activities performed at the Distribution department of Waterbedrijf Groningen?

To reach this goal, first the most important factors that influence the efficiency and effectiveness of the maintenance process need to be identified. With maintenance process is intended: the total value stream of the initiated maintenance work up until finished work. This process is described extensively in paragraph 3.1. Improving the situation must lead to costs reduction for the future. The use of resources is directly linked to the costs of maintenance. Therefore, the costs are an important criterion to establish a maintenance policy. Besides this, the availability and reliability are considered to be important criteria for an optimal policy in water distribution networks (Loung, 2005). These criteria also directly affect the service level given to the customer.

2.2

Definition efficiency and effectiveness

2.3

Methodology

To answer the research question, the “DDC” methodology developed by De Leeuw, is used as a structured approach to the problem situation. First, the problem situation is described and supported with a conceptual model of the current situation. An important aspect in the diagnosis phase is to look at the problem situation multiform which means

looking at the problem situation from different points of view. The analysis will clarify the causes of the main problem and will be the basis for improvements. Also, the relations between the main research objective and the factors that influence these have to become apparent. In this case explorative research is conducted. According to Blumberg, this kind of research is used

by researchers when there is no clear vision on the Fig. 4. DDC model (De Leeuw, 1996 )

possible obstruction they will be faced with or when the research area is new or unclear (Blumberg et al, 2005). The end of the diagnosis the most significant problem areas are clear and stated in an empirical model. The design consists of the actual proposed improvements at WBGR. Based on design criteria, the design of the potential improvements will be evaluated. Alternatives will be analyzed and the option that is most suitable in the situation will be chosen. The realization of the changes that are advised is carried over to WBGR.

This research can be characterized as a focused research to support the decision making for WBGR to increase performance of the maintenance process, applied for a specific context related to the maintenance activities performed by the OHD department. This has the purpose to investigate the influencing factors of efficiency, and explore causation in order to find underlying principles. Hence, this research can be considered as a case study. For that reason, the planning of this research is based on a five stage research process model (Stuart et. al., 2002).

Fig.5. The five stage research process model (Stuart et al., 2002)

The first phase in the research project is defining the problem statement and research questions. This consists of building a body of knowledge and developing theory. Ultimately, good theory is testable

D

DDiiiaaagggnnnooosssiiisss

C

and logically coherent, and can result in new, frame-breaking insights (Eisenhardt, 1989; Schmenner and Swink, 1998). For this research it will enhance discovery/ description/understanding; mapping variables/relationship building and finally theory validation/ extension/refinement.

The second step in conducting case research is the development of a research instrument and selection of the appropriate field sites. Having defined the research question, there is a need to develop measurement instruments to capture the data for future analysis. This will assist in providing focus: validity requirements and site selection such as volume and variety of research focus.

The third phase is data gathering. Ellram (1996) make a distinction between case studies by the type of data being gathered and the type of analysis applied in the research. The type of data gathered will concern empirical data from WBGR and both quantitative and qualitative analysis will be applied to come up with recommendations for WBGR. The “data” will be collected from face-to-face interviews, documents that the company is willing to provide (archives and previous research at WBGR), the researcher‟s observations and literature review such as research papers, books and internet. By means of interviews and a short questionnaire the factors that influence maintenance work will be tested and evaluated. The questionnaire will be performed for the mechanics to perceive an objective interpretation by rating of factors influencing work situations. One disadvantage is the missing factor of anonymity. For this reason it is important that the data that is gathered will be strictly confidential. Ten different mechanics will be chosen with different backgrounds and management interviewees will be chosen according to their responsibilities within the organization such as planners, team leaders or managers of different departments. Also, interviews will be taken at similar companies to analyze best practices. These interviews are based on half standardized interviews. This means that the subjects of the interviews are predetermined but the interviewee formulates their own answers and can influence the subjects with their own interpretations. Therefore open questions are asked. Because in this research it is important to determine true causal relationships, building the interviewee‟s trust quickly is vital for success (Stuart et. al., 2002). This approach for gaining the complete picture is consistent with the ethnographic data gathering approach and concepts of triangulation (Schein, 1987; Jick, 1979). This should provide validity within the case study.

The last phase, phase five, is dissemination of the research findings. This takes place by disseminate this thesis and by presentations at the company and the University of Groningen.

2.4

Sub questions

The sub questions will help to support the Main Research Question. By answering these questions, the topics about efficiency and effectiveness have to become apparent. Both quantitative and qualitative information will enhance the validity of the diagnosis of the problem situation. In order to lead to an answer of the main research question the sub questions are subsequently:

Diagnosis:

1 How can the current maintenance process be described and what activities are involved? 2 Which variables can be identified that can improve the efficiency and effectiveness of the

OHD department.

3 How can the performance of maintenance activities be measured?

4 Can best practices be found at comparable companies for procedures to plan maintenance? 5 Which factors need to be addressed that significantly improves the current situation? Design:

6 Which actions need to be taken to improve the factors that are concluded in the diagnosis? The diagnosis leads to an answer of sub question five. The design phase leads to conclusion and recommendations about the actions that need to be taken to improve the efficiency and effectiveness of the maintenance activities. It should be noted that sub question six is preliminary because outcomes of the diagnosis are unclear. Therefore sub-sub questions are formulated in this phase, that must lead to the actions that need to be taken.

2.5

Conceptual model

to be regarded as a complete picture of the situation. Significant factors that arise during the research will be taken into account. The model is divided between input factors and supporting factors. The input factors are given and determine the procedures of how to work. The supporting factors can be influenced and can potentially improve the situation for the OHD department. The system that needs to be investigated is considered to be all the maintenance activities performed by the maintenance mechanics of OHD and all the supporting administrative activities.

Motivation of employees Efficiency Planning and Scheduling Applicability of Inventory Control Capabilities of employees Quality of material and equipment

Efficiency and effectivenes of Maintenance function

Input factors Problem Supporting factors

Availability Employees Usability of CMMS Legislative rules Level of serviceability Quality of Cooperation Structure and condition of the

distribution network _{Maintenance costs}

Weather condition Customer information input Geographical obstacles Water pressure Balance of in-house capability and outsourced service Effectiveness Maintenance regimes

The main purpose of the maintenance is to continually serve the customer with a continuous flow of high quality water. Therefore the service level to the customer will be an important factor when looking at the efficiency of the department. Every task that is performed by the OHD department has the purpose to give service to the customer. More effort in maintenance brings higher service levels. For instance, if less mechanics are available, one can be more efficient in terms of costs of the department. However, the direct effect will be a decrease in service to the customer when, for example, several breakdowns occur that will increase the time to get to the customer because of a smaller amount of mechanics available. This has a negative impact on the customer satisfaction. Interruptions of water supply can occur due to system failures or planned maintenance. The frequency, timing and duration of interruptions can be affected by the capital investments and operations of the supplier, with higher levels of service generally being attainable through higher costs and hence higher prices (Hencher, 2004). To determine the appropriate level of these service attributes relative to price, information is needed on the value that customers place on each attribute. Therefore, also the serviceability has to be taken into account and is described in section 3.3. Next to the serviceability, the quality of work that is delivered by maintenance mechanics is an important selection criterion when designing improvements for WBGR in this research. The quality of work determines the effectiveness of the executed maintenance work. Therefore, the selection criteria used for this research are based on the appropriate mix of the serviceability, quality and costs.

2.5.1 Input factors

The input factors are explained as follows:

 Customer input factors are related to the complaints or advice of customers. This information exists of problems with water supply such as low water pressure or brown water. This is important information for WBGR to act upon. Urgency of complaints defines the reaction time to go to the location. As such, the number of complaints in a certain region defines the maintenance that has to be done. More complaints means worse condition of the distribution network and when this number exceeds a threshold of complaints the network can be considered for sanitation.

is higher delivery certainty. With the structure is also intended: the configuration of objects in the network. These are the number of valves in the network. When for example the number of valves are reduced, less maintenance is required. This will be elaborated in section 4.2.  The condition of the network is essential to prevent breakdowns. Older or corrosive sensitive

material is more subjected to failures than newer, often higher quality, material. Much research is executed to estimate the optimal time of water pipe renewal. In cooperation with asset management the condition is determined and modified as described in the section asset management.

 With geographical obstacles are meant: different types of soil and obstacles in the ground or on top of it. Thereby, the depth of the pipes influences the time to reach the pipes for maintenance mechanics.

 The weather condition is also a given factor that cannot be influenced. Typical weather types such as rain, frost or dry weather determine the difficulty to execute the work. Frost makes the soil very rigid and thus difficult to access. There is a seasonal change in number of network failures. In the summer the number of breakdowns is much lower than during the rest of the year.

 The availability of employees determines how much work can be dedicated to maintenance activities. This also depends on the time spent on other activities apart from the maintenance work. These are factors such as: free time, illness, meetings and availability of hired work. On the other hand, if there is no work for the mechanics it results in idle time that costs money.  The legislation is an important factor that needs to be taken into account when executing the

 Water pressure is mentioned because the pressure in the network can be directly related to the number of failures in the network. In research performed by Arajou a model is created to support decision systems regarding the quantification, location and opening adjustment of control valves in a network system, with the main objective to minimize pressures and consequently leakage levels (Arajou et. al. 2006).

2.5.2 Supporting factors

The supporting factors can potentially improve the situation. By examining these factors, the input factors are taken into account.

 More efficient inventory control systems, which are a part of the planning of maintenance, will allow mechanics to work more efficient. It can ensure that mechanics always have their materials ready for their activities and prevent delays in waiting time for parts, the extra time to drive to another storage place or delivering parts by other employees at the location. Next to this, an efficient inventory system can reduce cost significantly.

 The usability of equipment that is used for maintenance influences the efficiency to implement the material and the effectiveness of the material. Efficiency is gained when the time to repair is reduced. Also standardization in materials and equipment can help the efficiency of the maintenance. Effectiveness is reached when the work is executed the right way. For example, when the time that a repair or change in the network will remain operative. The quality of the material is essential to maintain the network in a good condition.

 To carry out an efficient and effective planning and scheduling the amount of work needs to be estimated and the different types of work need to be identified and prioritized. The planning process is interpreted as a decision making process aiming at the selection of appropriate actions that are to be carried out in practice such that the planned system will achieve certain objectives (Ackoff, 1984). An essential part of planning and scheduling is to forecast future work and to balance the workload between categories. There are roughly two different kinds of work. These are: planned maintenance and unplanned maintenance. In most of the cases the planner has to make a rough estimate of the duration of work. Thereby, a balance has to be found between planned and unplanned work.

 Motivation of employees will influence the performance of the maintenance activities. It is recognized that there is resistance to changes. This counts for mechanics as well as for the managers and the relation between these two. Management style is important. Changes in work processes can have influence on the motivation of employees. Increasing motivation can have a positive influence on productivity. Therefore motivation is an important factor that needs to be investigated.

 Capabilities of employees consist of the knowledge that maintenance mechanics possess to perform their work correctly. This is knowledge about work procedures, material use and knowledge about the area.

 Balance of in-house capability and outsourced service, if the capacity is insufficient or the in-house capabilities can be limited, WBGR can choose to outsource the maintenance work. This can be a cheaper solution when the work is specialized.

 Maintenance regimes are the policies on the strategic level and determines how the maintenance should be planned for the longer term. This entails, for example, the amount of preventive maintenance of the physical assets in the distribution network.

3. Diagnosis

To gain insight in the maintenance department and the organization in which it performs its activities, the first part of this chapter describes the maintenance processes, information flows and performance measurement. Thereafter, from paragraph 3.4, the supporting factors of the conceptual model will be analyzed and a selection will be made for improvement.

3.1

Maintenance process

This part of the thesis illustrates the maintenance process from the sources of incoming work until finished maintenance work. Figure 7. describes the work flows that ensue from the different sources and how this work will be processed.

OH-D Maintenenance proces W or k pl an ni ng O H -D W or k pl an ni ng O H -D A tte nt io n w or k pr ep ar at io n A tte nt io n w or k pr ep ar at io n F in is he d w or k F in is he d w or k S ta rt m ai nt en an ce w or k S ta rt m ai nt en an ce w or k D ep ar tm en t D ep ar tm en t W or k so ur ce W or k so ur ce

Customers Third parties

Work Order

Maintenance mechanics Attendence work

Special services

Schedule the work

Incompleted work

Finished

attendence Technical finished work

Subcontracters

Check workorders

Internal Projects Onderhoud OH-D

TKC NBD Production Planning OH-D

Incoming work at

planner OH-D Work preparation Decline the work

Administrative finished work

KLIC

The maintenance work originates from four different sources which are further explained in the sections below. The complete work packages will be processed and scheduled by the planners and given to the maintenance mechanics by means of a work-order. Often the work is assigned to more than one mechanic and work as a team on a project. The mechanics use a laptop computer to communicate and to help performing their jobs. Much communication is done via mail. After the work is finished, the maintenance mechanics send the work orders back to the planner (check work order). The hours spent by mechanics are filled in on paper. The planner processes the material that is used and the working hours spent on each tasks in Agresso for financial administration. The paper work-orders are thrown away afterwards which makes it difficult to look back on what, when and by whom the maintenance work is performed.

Customer Contacts orders

The orders that result from customer complaints are the first source. This kind of work is mostly requested by the TKC department. These are complaints from the customer like: failures in the network, water quality/pressure complaints or requests for changes in the network or water meters. When a customer has a complaint about water supply, the department TKC makes an appointment with the customer to solve this problem if needed. This can be from both customers and people who detected or caused failures. When there are complaints outside office hours, the shift service leaders processes the complaints received from Interswitch that receives the calls. For larger failures mostly subcontractors assist the mechanics.

Third parties

Secondly, work has to be done for third parties such as private individuals or local or provincial authorities. This work will be prepared by the engineering department (NBD). This is work that derived from projects in the province like: building new roads or housing. As a result, modifications of pipes have to be done when the roads are changed, new buildings are made or other activities for which water pipes have to be changed. Also changes have to be made when private individuals wants to do something in their property where a water network is located. Therefore WBGR always wants to build new networks in public ground. Most of this work is done by subcontractors. Smaller and more specialized work is done by the mechanics of WBGR. This work is planned at least more than 13 weeks in advance because of preparation of the work. It is prepared by NBD and involves the material preparations permissions, work drawings, and estimation and arrangement of the work that has to be done.

Internal Projects

The third source is internal projects. These are projects from the engineering department (NBD) department who are responsible for renovation and planning of new distribution networks, reconstructions and/or replacements. The work that ensued from projects are usually own initiated projects such as sanitations, changes in the network or new pipes. This is often planned far in advance by NBD. Most of this work is done by subcontractors. Some of the work is done by OHD. The orders arrive at the planners and team leaders of OHD with project maps and by mail or telephone for further agreements. Sanitations are mostly done in cooperation with municipalities. About 80% of the projects of WBGR originate from the local government. For example, when a sanitation of sewage network, WBGR prefers to combine sanitation of water pipes. This saves work and annoyance for the residents in the neighbourhood.

Initiated work of the OHD department

3.2

Maintenance of distribution assets

This paragraph concerns the maintenance types and how these types of work are planned. There are roughly three types of maintenance performed by OHD. These are regular maintenance, failures and work directly linked to the customer. The graph below shows the amount of productive work that is estimated for these types of work. The rest of the work is dedicated to WION and other management activities. The percentages are based on FTE assignment (subcontractors included) at OHD department. This data is derived from productive working hours written by mechanics.

Fig 8. Average costs during the year of the OHD department in 2009 (Administration OHD department)

Regular maintenance

Based on FTE allocation, roughly 20% of OHD resources are associated to this kind of maintenance. which consists of maintenance of fire monitors, valves, water meters, water pipe flushing or taking water samples. Fire monitors and valves are maintained often combined with each other and thereby combined with flushing activities. Currently, most of these activities are based on cyclic maintenance. It is currently being investigated by the department whether this is a smart way to work.

Another example is the maintenance of valves. The maintenance is reconsidered because the valves are checked every four years while checking all these valves is not really necessary. Hence, some more critical valves need to be indicated that can be checked more regularly. Next to this, the number of working valves can be reduced. It is stated that not all the valves are necessary which decrease the amount of work drastically. The direct disadvantage is that the sections will be larger and more people will be affected when sections need to be closed. The process of identifying the importance of valves and pipes is studied by Jun et. al. 2008, who analyzes two methodologies, called Pipe-by-Pipe or Valve-by-Valve, that are found to be a valuable tool for ensuring efficient operation and developing appropriate maintenance strategies, and thereby for improving the reliability of water distribution systems.

Customer work

Roughly 35% of resources of OHD are associated with customer work. This kind of work is chiefly described in the section customer contacts orders. It is limited scheduled, service focussed, image sensible and efficiency is strongly influenced by means of an effective planning. The planning of locations and times can strongly influence the efficiency to carry out the work.

Failures

3.3

Measuring performance of the maintenance work

To gain insight in efficiency there is a need for indicators to measure the performance of the OHD department. In order for a measurement system to be successful there is a need to know:

 How can the current level of efficiency be determined?  What is the desired level of efficiency in quantified terms?

 What are the influences of taking action to improve efficiency from the current level to the desired level of efficiency?

As already mentioned, the most important criteria for determining an effective maintenance policy are the maintenance costs and availability/reliability. This depends on the efficiency and effectiveness of the work that is performed. Tracking the performance of maintenance operations needs to be a key management issue when maintenance spending is a significant part of the operation budget (Tsang, 1999). The efficiency can be measured by looking at the input output model of the maintenance function as a black box. It can be measured for any input resource as the (theoretical) input divided by the actual input. When measuring the labour efficiency, the man-hours planned to a certain job divided by the actual man-hours needed. The effectiveness can be measured by the actual output by the theoretical output (Alsyouf, 2006).

For example, balance scorecards (BSC) provide a framework for translating the strategy into operational measures. The balanced scorecard allows the creation of KPIs for measuring maintenance management performance which are aligned to the strategic objectives. It gives the opportunity to look at the business from four different perspectives. These are: customer perspective, internal business perspective (processes), innovation and learning perspective (capabilities) and financial perspective (Kaplan and Norton, 2005). These four perspectives need to be in balance. Moreover, the balanced scorecard is a top-down approach, which means that the manager is responsible for translating the vision towards the employees of the company or department. The KPIs have to be the basis for achieving performance targets of the maintenance department (Marquez et. al. 2009).

maintenance decisions, scheduling and execution of work. The cycle is closed by capturing and analysis of performance data as shown in figure 9. In order to reach this, historical data needs to be collected about number and duration of maintenance work and costs. With this information, a more effective planning and control is possible for both short term and long term planning.

Fig 9. General model of management processes embedded in maintenance function (Tsang, 1999)

After the introduction of the periodic benchmark, companies are forced to look at performance indicators such as customer satisfaction, underutilized delivery minutes and costs. Service levels are determined by customer satisfaction surveys. To execute the strategy on operational level, goals need to be set. Therefore, a system is needed that measures the performance of the maintenance activities. Tsang (2002) has identified four strategic dimensions that are relevant for the maintenance function. These are:

1. The choice between in-house capability and outsourced service.

From the customer perspective of the BSC framework, the level of serviceability to the customer is an increasingly important factor that is incorporated in the company but also most difficult to measure. For serviceability, these indicators can be related to the speed, availability, and quality (Haarman, Delahay, 2003). The power of these indicators is that they are measurable and thus controllable. The following indicators that track the performance of serviceability are partly defined at WBGR and possible indicators ensued from literature are mentioned.

 Speed: Important factors that determine the speed related to the service to the customer are: response times to failures, repair times and flexibility in customer appointments. These factors are not measured within WBGR. When failures occur, the maximum time to repair is 24 hours (drinking water law). The time to reach a location of a failure is 1 hour. It is mostly based on „as soon as possible‟.

 Availability: Within the company, a measurement for availability is OLM‟s. These are minutes of interruption in water supply to the customer. OLM measurement is of increasing importance for assessment in the benchmark. This information is often used to determine the condition of the network compared to other companies rather than for customer service since it is questionable if there is a relation between OLM‟s and customer service. Because it is almost a usual fact that there is water, the customers don‟t consider it as a problem if there is no water for a while. Another measure of availability can be the availability of the material. An indicator can be the average spare parts inventory level. If material is temporarily not available when failures occur, it directly influence the service to the customer. The purpose is minimum number and duration of failures.

 Quality: These are the factors that influence the quality of work of maintenance and ways to communicate to the customer. Performance indicators can be the number of complaints about the customer contact, the mean time to repair assets or number of accidents.

3.4

Maintenance regimes

Maintenance regimes are decisions made for the tactical maintenance planning which is the in charge of Asset Management. This department is responsible for the condition and structure of the distribution network and makes decisions about renewal of pipes. The policy of Asset Management is focussed on delivering a high performance with low costs. This means that the quality and quantity of the water supply must be guaranteed for the future. Striking the important balance between issues such as risk and cost, or long-term growth and short-term gain is a part of the daily decisions that are required as a part of their decision making process. However in attempting to balance these issues the department is confronted with a different array of possible solutions to their problems. The decisions they make are based on criteria such as: quality of the pipes, impact on environment and number of households involved. Insight in the maintenance status of the pipes is essential in making proper decisions. Thereby, well underpinned criteria for modification of pipes will assist to make these decisions. These decisions are derived from risk analysis and central knowledge system of lifespan. With the assistance of these two methods the risk on the environment and the technical lifespan of the pipes can be determined.

The knowledge system of lifespan is a method that is developed by KWR and is used by the most of the water companies in the Netherlands. KWR has conducted a lot of research on determining the condition of pipes. They have developed a program to make sanitation decisions more objective, uniform and reproducible. The program exists of a technical deformation model, customer perception model and economic model. A technical deformation model estimates the current condition of the pipes with technical parameters. The customer perception model estimates the impact of failures on environment according to the perception of the customer. The economic model gives a prognosis of the trade-off between maintaining the pipes and the renewal of pipes. This way, the optimal sanitation moment can be predicted. At the moment this system is not fully incorporated at WBGR. Therefore information about failures, complaints and maintenance is essential for a fully working decision model.

In addition to this, strategies can be either a cost based (Kleiner et. al. 1998) approach or an availability/reliability approach (Li, Haimes, 1992). The cost based approach can be related to the whole life costing (WLC) methodology which has been developed for determining long term maintenance expenditure requirements for water distribution networks. When distribution networks include larger pipes with very high supply capacity and pipe breaks cause severe supply deficiency the use of availability/reliability seems more appropriate (Luong, 2005). According to this author, allocation of funds to different pipes depend on three main factors namely: (1) the importance levels of pipes in the network, (2) the maintenance cost rate required by the pipes in comparison with each other, (3) the individual optimal availability of each pipe in relation to the provided maintenance funds. It should be noted that these models can be limited due to the requirement of a vast amount of accurate input data which are often difficult to acquire.

The development of effective maintenance regimes is sufficient at WBGR. However, according the management of Asset Management the collection of information about maintenance is incomplete and can be more efficient. In the design phase will elaborate this subject on how this collection can be improved by means of CMMS.

3.5

Quality material and equipment

3.6

Balance of in-house capability and outsourced service

Concerning the choice of in-house capabilities and outsourcing of maintenance work, it is stated that the outsourcing of maintenance need to be minimized. For the mechanics it is arranged that all the work that is physically heavy or difficult to execute (due to specific knowledge or equipment) it the maintenance work will be outsourced. Also, when the capacity of the workforce is limited, the work will be outsourced. Therefore the goals that are set in the year planning have to be met. The potential gains in performance with balancing these two choices is not analyzed in this research. Therefore this aspect is not further explained.

3.7

Planning and problem solving

This paragraph describes the planning and scheduling of the maintenance work on the operational level. It is stated that effective planning and scheduling improves the maintenance effectiveness significantly (Pete, 2006). With improvement is intended: minimizing the idle time of maintenance mechanics, maximizing the efficient use of work time, material and equipment and maintaining the operating equipment at a responsive level to the need of the customer in terms of water supply and quality. Here, the urgency of the job has to be estimated for every situation by the planners. These factors largely influence the decisions to plan the work.

The purpose of planning is:

“The purpose of planning is to achieve a common understanding of the tasks, to obtain an overview of the work to be carried out, to lay the foundation for allocating and committing resources, to be able to form a suitable organisation of work and to define a programme of monitoring and control”

(Andersen, Grude and Haug, 1995).

A definition of planning at the higher levels of management is an analytical process which includes an assessment in the future and a determination of desired objectives in the context of the future (Thierauf, 1982). Hayes-Roth define planning as determining in advance a course of action aimed at achieving a specific goal. They regard planning as the first stage of a two stage problem solving process. The second stage in this process is control: monitoring and guiding of the plan to a successful conclusion.

the aspects of the problem situation that the decision maker cannot control are uncontrolled variables. These two aspects together constitute the problem environment, (4) constrains imposed from the possible values of the controlled and uncontrolled variables, (5) the possible outcomes produced jointly by the decision makers choice and the uncontrolled variables.

Problems can be regarded as well-structured or ill-structured (Verbreak, 1991). It is well-structured when it meets the following requirements and is ill-structured if it fails to meet one or more requirements:

- A set of action alternatives is finite and identifiable.

- The solution is consistently derived from a model that shows a good correlation with reality. - The effectiveness or the efficiency of the action alternatives can be definitely e.g. numerically,

evaluated.

Scheduling is an ill-structured problem which means that in most cases the set of alternatives is only partly known and the models that are used rarely show a good correlation with reality due to fluctuation and unforeseen changes. The planner of WBGR is also faced with these kinds of dilemmas. This makes it difficult to evaluate the effectiveness and efficiency of alternatives because unknown factors influence the realization of a schedule.

Planning can be categorized in a long range planning which cover a time period of three to five years and sets plans for future activities and long-range improvement, medium-range planning which covers a period of one month to one year that sets plans for balancing capacity needed for labour and material acquisition. The last category is short-range planning, often called scheduling, that covers a period of one day to one or two weeks and focuses on the determination of all the elements required to perform maintenance tasks in advance. Scheduling defined the timing and volumes of specific maintenance work where the key decision variables are the start time, it‟s duration and volume. Based on criticality of the jobs, a job priority system can be applicable for short range planning (Hist and Hawkins, 2008). A recommended classification is: emergency needs to be solved immediately, Urgent needs to be solved within a few hours, Critical needs to be solve within 24 hours, Rush must be performed within end of current week and Essential but deferrable, before end of the next week. A similar division is used at WBGR, but no clear classification is made. Also for maintenance jobs with high variability are planned first on a day in order to minimize delays at the end of the day.