Introduction to the Theory and Application of Data Envelopment Analysis : A Foundation Text with Integrated Software

List of Tables xiii List of Figures xv Preface xvii Abbreviations xxii Introduction to Performance Measurement 1(20) Introduction 1(1) Why Measure Performance? 1(3) Performance Measurement Methods in Outline 4(9) Unit of assessment 4(1) Performance Indicators 5(1) Modelling Methods of Comparative Performance Measurement 6(1) Parametric methods for measuring comparative performance 6(3) Non-Parametric Methods for Measuring Comparative Performance 9(4) Some Areas Where Uses of DEA Have Been Reported 13(6) Financial Services 13(2) Regulation 15(2) Police Services 17(2) Conclusion 19(2) Definitions of Efficiency and Related Measures 21(16) Introduction 21(1) Unit of Assessment and Input-Output variables 21(1) Pareto-efficiency and Measures of Input and Output Efficiency 22(4) Input Overall, Allocative and Technical Efficiencies 26(3) An Illustration 29(2) Questions 31(6) Appendix 2.1: Mathematical Definitions 32(2) Appendix 2.2: Deriving Graphical Measures of Input Efficiencies 34(3) Data Envelopment Analysis Under Constant Returns to Scale: Basic Principles 37(28) Introduction 37(1) Basic Steps in Measuring Efficiency by DEA 37(4) Constructing a Production Possibility Set in the Single-input Single-output Case 38(2) Using the PPS to Derive Efficiency Measures 40(1) Using Linear Programming to Measure Efficiency in the Single-input Single-output Case 41(4) Using DEA to Measure Technical Input Efficiency in the Single-Output Multi-input Case: A Graphical Illustration 45(4) Using Linear Programming to Measure Technical Input Efficiency in the Single-output Multi-input Case 49(3) Using DEA to Measure Technical Output Efficiency in the Single-input Multi-output Case: A Graphical Illustration 52(2) Using Linear Programming to Measure Technical Output Efficiency in the Single-input Multi-Output Case 54(3) Questions 57(8) Appendix 3.1: Introduction to Linear Programming 59(5) Appendix 3.2: Postulates for Constructing the Production Possibility Set in DEA Under Constant Returns to Scale 64(1) Data Envelopment Analysis Under Constant Returns to Scale: General Models 65(24) Introduction 65(1) A General Purpose Linear Programming Model for Assessing Technical Input Efficiency 65(3) A General Purpose Linear Programming Model for Assessing Technical Output Efficiency 68(3) Value-Based DEA Models 71(3) Interpretation of Value-Based DEA Models 74(4) Efficient Peers and Targets in DEA 78(3) Targets 78(2) Efficient Peers 80(1) Input and Output Allocative Efficiencies 81(4) Cost Minimising Efficiencies 81(2) Revenue Maximising Efficiencies 83(2) Questions 85(4) Using Data Envelopment Analysis in Practice 89(34) Introduction 89(1) Choosing Inputs and Outputs in A DEA Assessment 89(3) Information Obtained in the Course of a DEA Assessment 92(1) Interpreting the Solution of a DEA Envelopment Model 93(5) Pareto-Inefficient DMUs 94(3) Pareto-Efficient DMUs 97(1) Interpreting the Solution of a Value-based Dea Model 98(7) Pareto-Efficient DMUs 99(4) Pareto-Inefficient DMUs 103(2) An Illustrative Use of Warwick DEA Software 105(6) Practical Tips for Carrying Out DEA Assessments 111(3) Questions 114(9) Appendix 5.1: Warwick DEA Software 119(1) Appendix 5.2 120(3) Data Envelopment Analysis Under Variable Returns to Scale 123(40) Introduction 123(1) The Concept of Returns to Scale 124(1) Assessing DEA Efficiency Under Variable Returns to Scale: A Graphical Illustration 125(4) Assessing DEA Efficiency Under Variable Returns to Scale: The Generic Envelopment Model 129(7) Value-based DEA Models Under VRS 136(3) Scale Efficiency, Returns to Scale and Most Productive Scale Size 139(14) Scale Efficiency 140(2) Identifying Returns to Scale By Means of DEA Models 142(5) Most Productive Scale Size 147(6) Practical Use of DEA Models Under VRS 153(2) Envelopment Models 153(1) Value-based Models 154(1) DEA Assessments Under VRS By Means of Warwick DEA Software 155(3) DEA Assessments Under Non-Increasing or Non-Decreasing Returns to Scale 158(2) Questions 160(3) Assessing Policy Effectiveness and Productivity Change Using DEA 163(36) Introduction 163(1) Disentangling Managerial and Policy Efficiency: An outline 164(4) Disentangling Managerial and Policy Efficiency: The generic approach 168(3) Disentangling Managerial and Policy Efficiency: Illustrative examples 171(4) Assessing Productivity Change By Means of DEA: A Graphical introduction to the Malmquist Index 175(6) Assessing Productivity Change By Means of DEA: The Malmquist Index in the general case 181(3) Assessing Productivity Change Using Malmquist Indices: Illustrative example 184(5) Capturing the Impact of Scale Size Changes on Productivity 189(2) The Cost Malmquist Type Index 191(3) Questions 194(5) Appendix 7.1 196(3) Incorporating Value Judgements in DEA Assessments 199(28) Introduction 199(1) Why we May Want to Incorporate Value Judgements in DEA Assessments 199(2) Methods for Incorporating Value Judgements in DEA 201(1) Using Weights Restrictions to Incorporate Value Judgements in DEA 201(3) Restrictions Applied to DEA Weights 202(1) Restrictions Applied to Virtual Inputs and Outputs 203(1) Some Approaches to Estimating the Parameters of Weights Restrictions 204(5) Interpreting the Results of DEA Models with Weights Restrictions 209(3) Effects of Weights Restrictions on the Interpretation of the Efficiency Measure Yielded By a DEA Model 210(1) Effects of Weights Restrictions on DEA Targets 211(1) Additional Effects of Absolute Weights Restrictions 212(1) Using unobserved DMUs to incorporate Value Judgements in DEA 212(7) A Procedure for Improving Envelopment in DEA 214(3) Advantages and Drawbacks of UDMUs in Incorporating Value Judgements in DEA 217(2) Questions 219(8) Appendix 8.1: Some Features of the DEA Efficiency Measure Under Weights Restrictions 222(3) Appendix 8.2: Some Features of Targets Yielded by Dea Models Under Weights Restrictions 225(2) Extensions to Basic DEA Models 227(24) Introduction 227(1) Assessing Efficiency Under Exogenously Fixed Input-Output Variables 227(8) Identifying Preferred Pareto-Efficient Input-Output Levels by DEA 235(9) Case Where no Input or Output May Deteriorate 236(2) Case Where Some Input or Output May Deteriorate 238(6) Assessing Dmus in the Presence of Categorical Variables 244(4) Questions 248(3) A Limited User Guide for Warwick DEA Software 251(20) Introduction 251(1) Preparing your Data Input before you Invoke the Programme 252(1) Initiating a Run of the Programme 253(1) Options Menu 253(5) The Run Menu 258(7) The Advanced Menu 265(2) File Handling 267(1) Some Common Error Messages 268(3) Author Index 271(2) Topic Index 273(4) References 277