Advances in DEA Theory and Applications

KAORU TONE is with the National Graduate Institute for Policy Studies, Japan. His contribution to DEA has a variety of attainments. He authored a classical book Data Envelopment Analysis: A Comprehensive Text with Models, Applications, References and DEA-Solver Software under the co-authorship with Professor Cooper (University of Texas) and Professor Seiford (University of Michigan). He also published many papers on DEA in international journals. Kaoru Tone opened a new avenue for performance evaluation, called Slacks-based Measure (SBM) that is widely utilized over the world. His recent innovations include Network SBM, Dynamic SBM, Dynamic DEA with Network Structure, Congestion, Returns-to-growth in DEA, Ownership-specified Network DEA, Non-convex Frontier DEA, Past-Present-Future Inter-temporal DEA, Resampling DEA and SBM-Max.

• List of Contributors xxAbout the Authors xxiiPreface xxxiiPart I DEA Theory 11 Radial DEA Models 3Kaoru Tone1.1 Introduction 31.2 Basic Data 31.3 Input-Oriented CCR Model 41.4 The Input-Oriented BCC Model 61.5 The Output-Oriented Model 71.6 Assurance Region Method 81.7 The Assumptions Behind Radial Models 81.8 A Sample Radial Model 8References 102 Non-Radial DEA Models 11Kaoru Tone2.1 Introduction 112.2 The SBM Model 122.3 An Example of an SBM Model 152.4 The Dual Program of the SBM Model 172.5 Extensions of the SBM Model 172.6 Concluding Remarks 18References 193 Directional Distance DEA Models 20Hirofumi Fukuyama and William L. Weber3.1 Introduction 203.2 Directional Distance Model 203.3 Variable-Returns-to-Scale DD Models 233.4 Slacks-Based DD Model 233.5 Choice of Directional Vectors 25References 264 Super-Efficiency DEA Models 28Kaoru Tone4.1 Introduction 284.2 Radial Super-Efficiency Models 284.3 Non-Radial Super-Efficiency Models 294.4 An Example of a Super-Efficiency Model 31References 325 Determining Returns to Scale in the VRS DEA Model 33Biresh K. Sahoo and Kaoru Tone5.1 Introduction 335.2 Technology Specification and Scale Elasticity 345.3 Summary 37References 376 Malmquist Productivity Index Models 40Kaoru Tone and Miki Tsutsui6.1 Introduction 406.2 Radial Malmquist Model 436.3 Non-Radial and Oriented Malmquist Model 456.4 Non-Radial and Non-Oriented Malmquist Model 476.5 Cumulative Malmquist Index (CMI) 486.6 Adjusted Malmquist Index (AMI) 496.7 Numerical Example 506.8 Concluding Remarks 55References 557 The Network DEA Model 57Kaoru Tone and Miki Tsutsui7.1 Introduction 577.2 Notation and Production Possibility Set 587.3 Description of Network Structure 597.4 Objective Functions and Efficiencies 61Reference 638 The Dynamic DEA Model 64Kaoru Tone and Miki Tsutsui8.1 Introduction 648.2 Notation and Production Possibility Set 658.3 Description of Dynamic Structure 678.4 Objective Functions and Efficiencies 698.5 Dynamic Malmquist Index 71References 739 The Dynamic Network DEA Model 74Kaoru Tone and Miki Tsutsui9.1 Introduction 749.2 Notation and Production Possibility Set 759.3 Description of Dynamic Network Structure 779.4 Objective Function and Efficiencies 809.5 Dynamic Divisional Malmquist Index 82References 8410 Stochastic DEA: The Regression-Based Approach 85Andrew L. Johnson10.1 Introduction 8510.2 Review of Literature on Stochastic DEA 8710.3 Conclusions 96References 9611 A Comparative Study of AHP and DEA 100Kaoru Tone11.1 Introduction 10011.2 A Glimpse of Data Envelopment Analysis 10011.3 Benefit/Cost Analysis by Analytic Hierarchy Process 10211.4 Efficiencies in AHP and DEA 10411.5 Concluding Remarks 105References 10612 A Computational Method for Solving DEA Problems with Infinitely Many DMUs 107Abraham Charnes and Kaoru Tone12.1 Introduction 10712.2 Problem 10812.3 Outline of the Method 10912.4 Details of the Method When Z is One-Dimensional 11012.5 General Case 11312.6 Concluding Remarks (by Tone) 115Appendix 12.A Proof of Theorem 12.1 115Appendix 12.B Proof of Theorem 12.2 116Reference 116Part II Dea Applications (past–present Scenario) 11713 Examining the Productive Performance of Life Insurance Corporation of India 119Kaoru Tone and Biresh K. Sahoo13.1 Introduction 11913.2 Nonparametric Approach to Measuring Scale Elasticity 12113.3 The Dataset for LIC Operations 12813.4 Results and Discussion 13013.5 Concluding Remarks 136References 13614 An Account of DEA-Based Contributions in the Banking Sector 141Jamal Ouenniche, Skarleth Carrales, Kaoru Tone and Hirofumi Fukuyama14.1 Introduction 14114.2 Performance Evaluation of Banks: A Detailed Account 14214.3 Current State of the Art Summarized 15414.4 Conclusion 163References 16915 DEA in the Healthcare Sector 172Hiroyuki Kawaguchi, Kaoru Tone and Miki Tsutsui15.1 Introduction 17215.2 Method and Data 17415.3 Results 18415.4 Discussion 188Acknowledgements 189References 19016 DEA in the Transport Sector 192Ming-Miin Yu and Li-Hsueh Chen16.1 Introduction 19216.2 DNDEA in Transport 19416.3 Extension 20016.4 Application 20716.5 Conclusions 212References 21217 Dynamic Network Efficiency of Japanese Prefectures 216Hirofumi Fukuyama, Atsuo Hashimoto, Kaoru Tone and William L. Weber17.1 Introduction 21617.2 Multiperiod Dynamic Multiprocess Network 21717.3 Efficiency/Productivity Measurement 22117.4 Empirical Application 22217.5 Conclusions 229References 22918 A Quantitative Analysis of Market Utilization in Electric Power Companies 231Miki Tsutsui and Kaoru Tone18.1 Introduction 23118.2 The Functions of the Trading Division 23218.3 Measuring the Effect of Energy Trading 23518.4 DEA Calculation 24218.5 Empirical Results 24318.6 Concluding Remarks 248References 24919 DEA in Resource Allocation 250Ming-Miin Yu and Li-Hsueh Chen19.1 Introduction 25019.2 Centralized DEA in Resource Allocation 25219.3 Applications of Centralized DEA in Resource Allocation 26119.4 Extension 26519.5 Conclusions 268References 26820 How to Deal with Non-convex Frontiers in Data Envelopment Analysis 271Kaoru Tone and Miki Tsutsui20.1 Introduction 27120.2 Global Formulation 27320.3 In-cluster Issue: Scale- and Cluster-Adjusted DEA Score 27620.4 An Illustrative Example 28120.5 The Radial-Model Case 28420.6 Scale-Dependent Dataset and Scale Elasticity 28720.7 Application to a Dataset Concerning Japanese National Universities 28920.8 Conclusions 294Appendix 20.A Clustering Using Returns to Scale and Scale Efficiency 295Appendix 20.B Proofs of Propositions 295References 29821 Using DEA to Analyze the Efficiency of Welfare Offices and Influencing Factors: The Case of Japan’s Municipal Public Assistance Programs 300Masayoshi Hayashi21.1 Introduction 30021.2 Institutional Background, DEA, and Efficiency Scores 30121.3 External Effects on Efficiency 30421.4 Quantile Regression Analysis 30921.5 Concluding Remarks 312Acknowledgements 312References 31222 DEA as a Kaizen Tool: SBM Variations Revisited 315Kaoru Tone22.1 Introduction 31522.2 The SBM-Min Model 31622.3 The SBM-Max Model 31822.4 Observations 32122.5 Numerical Examples 32322.6 Conclusions 330References 330Part III Dea for Forecasting and Decision-making (past–present–future Scenario) 33123 Corporate Failure Analysis Using SBM 333Joseph C. Paradi, Xiaopeng Yang and Kaoru Tone23.1 Introduction 33323.2 Literature Review 33423.3 Methodology 34023.4 Application to Bankruptcy Prediction 34323.5 Conclusions 352References 35424 Ranking of Bankruptcy Prediction Models under Multiple Criteria 357Jamal Ouenniche, Mohammad M. Mousavi, Bing Xu and Kaoru Tone24.1 Introduction 35724.2 An Overview of Bankruptcy Prediction Models 35924.3 A Slacks-Based Super-Efficiency Framework for Assessing Bankruptcy Prediction Models 36624.4 Empirical Results from Super-Efficiency DEA 37224.5 Conclusion 376References 37725 DEA in Performance Evaluation of Crude Oil Prediction Models 381Jamal Ouenniche, Bing Xu and Kaoru Tone25.1 Introduction 38125.2 An Overview of Crude Oil Prices and Their Volatilities 38525.3 Assessment of Prediction Models of Crude Oil Price Volatility 38825.4 Conclusion 401References 40226 Predictive Efficiency Analysis: A Study of US Hospitals 404Andrew L. Johnson and Chia-Yen Lee26.1 Introduction 40426.2 Modeling of Predictive Efficiency 40526.3 Study of US Hospitals 40826.4 Forecasting, Benchmarking, and Frontier Shifting 41226.5 Conclusions 416References 41727 Efficiency Prediction Using Fuzzy Piecewise Autoregression 419Ming-Miin Yu and Bo Hsiao27.1 Introduction 41927.2 Efficiency Prediction 42027.3 Modeling and Formulation 42327.4 Illustrating the Application 43327.5 Discussion 43827.6 Conclusion 440References 44128 Time Series Benchmarking Analysis for New Product Scheduling: Who Are the Competitors and How Fast Are They Moving Forward? 443Dong-Joon Lim and Timothy R. Anderson28.1 Introduction 44328.2 Methodology 44528.3 Application: Commercial Airplane Development 44928.4 Conclusion and Matters for Future Work 454References 45529 DEA Score Confidence Intervals with Past–Present and Past–Present–Future-Based Resampling 459Kaoru Tone and Jamal Ouenniche29.1 Introduction 45929.2 Proposed Methodology 46129.3 An Application to Healthcare 46529.4 Conclusion 476References 47830 DEA Models Incorporating Uncertain Future Performance 480Tsung-Sheng Chang, Kaoru Tone and Chen-Hui Wu30.1 Introduction 48030.2 Generalized Dynamic Evaluation Structures 48230.3 Future Performance Forecasts 48430.4 Generalized Dynamic DEA Models 48730.5 Empirical Study 49530.6 Conclusions 513References 51431 Site Selection for the Next-Generation Supercomputing Center of Japan 516Kaoru Tone31.1 Introduction 51631.2 Hierarchical Structure and Group Decision by AHP 51931.3 DEA Assurance Region Approach 52131.4 Application to the Site Selection Problem 52231.5 Decision and Conclusion 527References 527Appendix A: Dea-solver-pro 529Index 535