Methodology for warranty optimization and negotiation with specific focus on reliability and warranty attractiveness

A METHODOLOGY FOR WARRANTY OPTIMIZATION AND NEGOTIATION WITH SPECIFIC FOCUS ON RELIABILITY AND WARRANTY ATTRACTIVENESS

by

AMBAD PRASHANT MAHADEV

Submitted

in fulfillment of the requirements of the degree of

Doctor of Philosophy

to the

INDIAN INSTITUTE OF TECHNOLOGY DELHI

OCTOBER 2012

या प्रबंधातील संशोधन कायर् ,

ितथर् प

ी . महादेव भागुजी अंबड व

सौ . तुळसाबाई महादेव अंबड यांना समिपर्त.

i

ii

CERTIFICATE

This is to certify that the thesis entitled “A Methodology for Warranty Optimization and Negotiation with Specific Focus on Reliability and Warranty Attractiveness” being submitted by Ambad Prashant Mahadev to the Indian Institute of Technology Delhi for the award of the degree of Doctor of Philosophy is a record of original research work carried out by him. He has worked under my guidance and supervision and has fulfilled the requirements for the submission of this thesis, which to my knowledge has reached the requisite standard.

The results contained in this thesis have not been submitted, in part or full, to any other university or institute for the award of any degree or diploma.

(Dr. Makarand S. Kulkarni)

Associate Professor,

Department of Mechanical Engineering, Indian Institute of Technology Delhi,

New Delhi-110016, India

iii

ACKNOWLEDGEMENT

I express my heartiest thanks to Dr. Makarand S. Kulkarni whose motivation, guidance and encouragement drives me to successfully complete my research work. He created interest in the research in such a way that I did not realize my four years of stay at IIT. His continuous support helps me to come out of the difficult situations in the research problems. The leanings that I have achieved with him in this research build my confidence about capabilities to do the independent research in the future. His ethics made him a role model for me and show me a practical way to look at the life. I am greatly inspired by his communication, drafting and convincing skills. His presentations and enthusiasm in class will really help me in building the career path in my life.

I feel heartfelt gratitude and affection to my family and relatives for their extremely caring support during my research work. My life partner Vaishali given me constant back up, moral support and took care of parents during my rigour work of research. She always motivated me and takes care of me. I express my unconditional love to my son Pranit and feel extremely sorry for missing golden moments of his life. I would like to thank my sisters and my in-laws for their valuable support during my stay at IIT.

I am thankful to Head, Department of Mechanical Engineering, IIT Delhi for allowing me to carry out research and also providing me the support and facilities as and when needed.

I am thankful to R & D, Greaves Cotton Limited, Aurangabad for collaborating on a joint research project with IIT Delhi and extended their technical help in terms of expertise and data required for the research work. I am thankful to Mr. D.S. Gunjegaonkar for his encouragement without which this collaboration was not possible. I extend my thank to Mr. A.S. Tiwari and Mr. Rajiv Kulshreshtha for giving me the time for discussion and providing all type of help for the research work.

I express my deep sense of gratitude to Prof. O.P. Gandhi, ITMMEC, IIT Delhi for his constant encouragement and helping me as a Student Review Committee (SRC) member to

iv

make the report technically sound. I am immensely grateful to Prof. A.D. Gupta (Chairman, SRC) and Dr. A.K. Darpe (Member, SRC) for their valuable and constructive comments which helped in bringing completeness in my research.

I pay my sincere homage to Late Prof. Arun Kanda and sincere thanks to Dr. Kiran Seth under whom I achieved my lifetime experience while working as a TA. they showed me a different way of living the life.

I acknowledge my gratitude to Prof. S.G. Deshmukh whose valuable support helped us to form the Explorer group of researchers and carry out various research activities under his patronship. I am thankful to Dr. Vipul Jain and Dr. Nomesh Bolia for their friendly interactions from time to time and their valuable suggestions during my research work.

I am thankful to Dr. Bhupesh Lad and Dr. Divya Pandey who helped me from time to time and support me in every phase of my research work. I extend my thanks to Mr. Dipak Nehete and Mr. Pankaj Zine whose constant moral support helped me a comfortable stay away from my family. They were with me like a shadow during the entire stay at the IIT and supported me during my difficult times. I would also like to thank Mr. Md. Asjad, Mr. Vedpal and Mr. Pravin Tambe for their help and cooperation.

I will definitely miss the pleasant moments which I have spent with my colleagues and friends in the campus. I extend my sincere thanks to Dr. Rahul Mulik, Dr. A. C. Shukla, Dr. Anuj Prakash, Mr. Avinash Samvedi, Mr. Gajanan Panchal, Mr. Bhagatsingh Patil, Mr. Suhas Vasaikar, Mr. Subhash Lahane, Mr. Manoj Satyarthi and Dr. Raman Sharma for their friendly support. I am also thankful to Mr. Umang Soni, Mr. Amit Upadhay, Mr. Sumit Sakhuja, Mr. Anirban Kundu and Ms. Monica Tanwar for their valuable support.

I am thankful to Mr. Srichand Sharma and Mr. M.K. Bhatnagar who provided all the facilities in the Lab whenever needed by me. Last but not least I thank all those who helped me directly or indirectly in my research work.

(Ambad Prashant Mahadev)

ABSTRACT

In the purchase decisions of a product, buyers typically compare characteristics of comparable models of different brands available in the market. When brands available in the market are similar in terms of features, price and quality, the customer tend to choose the products based on the post sales support offered by the manufacturer. In this, warranty is one the most influential factors known to the buyer at the time of purchase. Attracting more customers as compared to the competitors with better post sales support is a challenging task and depends upon the terms of warranty policy and support level provided by the manufacturer. As the cost of warranty directly affects the profit margin of the manufacturer, the decisions regarding the terms of warranty play an important role at the design stage. In some cases, the price of the product is dictated by the customer; thus, achieving the minimum acceptable selling price with the targeted profit is also one of the important aspects which the manufacturer needs to consider.

In the present study, a methodology for warranty optimization and negotiation with specific focus on reliability and warranty attractiveness is developed. The methodology is developed by building models in order to address various issues related to warranty.

While designing a warranty policy, it would be beneficial to consider the extent to which a policy would appear attractive to the customer. This will be helpful in influencing the customer’s buying decision. However, the manufacturer needs to consider a number of effects while taking decisions related to warranty. For example, the attractiveness of the warranty will increase if the manufacturer provides a longer warranty duration but this will also increase the warranty cost. In order to minimize the warranty cost, reliability is one of the most important characteristics as the warranty cost strongly depends upon the reliability performance of the product. An increase in the reliability will most likely increase the

v

product cost but result in reduction in the number of failures which in turn will decrease the warranty cost and increase the customer satisfaction. In many cases, the manufacturer can identify component alternatives with different cost and reliability. Thus, a proper choice of component alternatives along with an appropriate warranty policy can be made to achieve the required level of attractiveness at the lowest possible cost.

In addition to parameters like cost of the component, duration of the warranty policy and the type of warranty policy, the warranty attractiveness will also be influenced by the support provided in the case of failure, which will reduce the inconvenience caused to the customer.

As a strategy, the manufacturers need to ensure that warranty attractiveness of their products is greater than that of the competitors. To be able to do this, a quantitative measure of warranty attractiveness is required so that by studying the attractiveness of the competitor’s warranty, an appropriate warranty policy can be selected. While doing so, the warranty cost also needs to be kept in mind.

To capture the attractiveness of the warranty from the customer’s point of view, a Warranty Attractiveness Index (WAI) is proposed, which comprises of three indices namely, Cost Criticality Index (CCI) which reflects the perception of a customer towards cost of the components, Policy Attractiveness Index (PAI) which reflects the perception of a customer towards the type and duration of the warranty policy and Support Attractiveness Index (SAI) which reflects the perception of a customer towards the support level available in the case of a failure. Further, an Attractiveness Advantage Ratio (AAR) is proposed which is used for comparing the overall warranty attractiveness with that of the competitor’s product.

Based on the attractiveness index, a warranty cost optimization approach is developed with component alternatives, warranty policy, warranty duration and the level of support as the decision variables. The objective is to minimize the warranty cost with specific focus on reliability and warranty attractiveness while meeting certain performance constraints.

vi

The manufacturer of the product would be also interested in achieving a number of other objectives while designing the warranty. However, it may be necessary to prioritize the order in which the objectives are achieved from a practical view point. For such an approach, it appeared more convenient to model it using goal programming. A multi-objective optimization model based on goal programming approach is developed. The multiple objectives are in the form of goals to be achieved with respect to warranty attractiveness, warranty cost, spare parts cost to the end user over the useful life of the product and the mean time between failures. The relative importance of the goals is incorporated by using weights for the goals.

Warranty is one of the tools used by the manufacturers to attract the customers, with a challenge of achieving this with a competitive selling price. When the product is a sub assembly supplied to the Original Equipment Manufacturer (OEM) who is the customer, decision regarding selling price is influenced heavily by the customer.

In the approaches proposed so far, the product price was not considered during optimization.

However, it may so happen that the product cost after optimization may turn out to be such that the manufacturer may not be able to make the desired profit at the price dictated by the buyer. In such cases, the deal between supplier and customer may get cancelled. However, instead of cancelling the deal, there may be a possibility of negotiation between the manufacturer and the customer. Modelling such a negotiation mechanism is a challenge as it will involve re-optimizing all the decision variables under varying degrees of relaxation of the target values set by the customer for many objectives.

A warranty based bilateral automated multi-issue negotiation model is developed in which the negotiation is carried out between the manufacturer and the OEM, where the manufacturer supplies the automobile engine to the OEM. The proposed negotiation approach

vii

viii

will help the manufacturers to take appropriate decisions related to the selling price considering the warranty parameters and other objectives at the design stage of the product.

In order to study the effect of variability of the input parameters, a variance propagation model is proposed which estimates the variance in a dependent or output variable as a function of the variance of the independent or input variables. Using this model, a probabilistic approach for warranty optimization is presented with an objective to minimize the warranty cost.

The solutions to the warranty optimization approaches are carried out using a genetic algorithm with elitism strategy. In the present research work, the approaches developed for the warranty optimization are demonstrated using a real life case of an automobile engine manufacturer.

In essence, the present research results into a systematic development of methodology to optimize the warranty considering the critical parameters that affect the warranty decisions at the design stage. The proposed models are useful to assess the attractiveness of the manufacturer’s warranty as compared to that of the competitor and will also help in warranty based negotiations.

TABLE OF CONTENTS

Dedication i Certificate ii Acknowledgement iii-iv

Abstract v-viii

Table of Contents ix-xiii

List of Figures xiv

List of Tables xv-xvi

List of Abbreviations and Notations xvii-xxi

CHAPTER 1: INTRODUCTION 1-15

1. 1 Preamble 1

1.1.1 Warranty Policies 2

1.1.2 Warranty Cost 4

1.1.3 Reliability 5

1.2 Motivation for the Present Work 6

1.3 Research Problem 8

1.4 Objectives of the Research 8

1.5 Research Methodology 9

1.6 Thesis Outline 13

1.7 Summary 15

CHAPTER 2: LITERATURE REVIEW 17-54

2.1 Introduction 17

2.2 Review Papers 19

2.3 Product Warranty 21

ix   

2.3.1 Warranty Policies 22

2.3.2 Warranty Cost 28

2.4 Warranty and Reliability 35

2.5 Customer Satisfaction 42

2.6 Problem Solution Approaches 44

2.7 Negotiation 46

2.8 Observations from the Literature Review 50

2.9 Research Gaps 53

2.10 Summary 54

CHAPTER 3: AN ATTRACTIVENESS INDEX BASED APPROACH FOR WARRANTY OPTIMIZATION

57-85

3.1 Introduction 57

3.2 Warranty Attractiveness Index 59

3.2.1 Cost Criticality Index 59

3.2.2 Policy Attractiveness Index 61

3.2.3 Support Attractiveness Index 64

3.3 Warranty Optimization Problem Formulation 69

3.3.1 Overall Attractiveness Index 71

3.3.2 Parameters Estimation for Failure Distribution 73

3.3.3 Warranty Cost Models 75

3.3.4 Mean Time Between Failures 77

3.3.5 Objective Function 78

3.4 Warranty Optimization 78

3.4.1 Steps for Genetic Algorithm 79

3.4.2 Optimization Results 81

x   

3.5 Summary 83 CHAPTER 4: GOAL PROGRAMMING APPROACH FOR MULTI -

OBJECTIVE WARRANTY OPTIMIZATION

87-113

4.1 Introduction 87

4.2 Warranty Optimization Problem Formulation 88

4.2.1 CRITIC Method 89

4.2.2 Goal Programming Model 102

4.2.2.1 Formulating the Objectives 103

4.2.2.2 Setting the Goals 104

4.2.2.3 Loss Functions for Objectives 105

4.2.2.4 Deviation Functions 107

4.2.2.5 Formulating the Objective Function 108

4.3 Warranty Optimization 109

4.4 Summary 112

CHAPTER 5: WARRANTY BASED NEGOTIATION 115-146

5.1 Introduction 115

5.2 Automated Negotiation Model 117

5.2.1 Negotiation Protocol 117

5.2.2 Negotiation Strategy 118

5.3 Warranty Based Automated Negotiation Process 120

5.3.1 Iteration 1: Setting of Goals 120

5.3.1.1 Stage 1: OEM side Activities 120 5.3.1.2 Stage 2: Manufacturer side Activities 120 5.3.2 Iteration 2: Setting Relaxation for Negotiation 121

5.4 Problem Formulation 123

xi   

5.5 Warranty Based Optimization Model 125 5.5.1 Loss Functions and Deviation Functions for Objectives 126

5.5.2 Objective Function 127

5.6 Negotiation Methodology 128

5.6.1 Illustration Case 1 129

5.6.2 Illustration Case 2 137

5.7 Summary 143

CHAPTER 6 :VARIANCE PROPAGATION MODEL 149-166

6.1 Introduction 149

6.1.1 The Variance Propagation Concept 149 6.2 Variance Propagation Model for Overall Attractiveness Index 151 6.2.1 Model for the Cost Criticality Index 152 6.2.2 Model for Policy Attractiveness Index 153 6.2.3 Model for Support Attractiveness Index 155 6.2.4 Calculations for Variance Propagation Model 159 6.3 Probabilistic Approach for Warranty Optimization 163

6.4 Summary 165

CHAPTER 7: CONCLUSION 167-174

7.1 Summary 167

7.2 Research Contributions 171

7.3 Utility of the Research Work 172

7.4 Limitations of the Study 172

7.5 Scope for Future Work 173

REFERENCES 175-204

APPENDICES 205-211

xii   

xiii   

APPENDIX A1: Mean values used for various input parameters of CCI, PAI and SAI

205

APPENDIX A2: List of Publications from research 209 APPENDIX A3: Biography of Researcher 211