Extension of data envelopment analysis with preference information. Value efficiency (Q905589)

This book introduces an approach of data envelopment analysis which is based on the idea to combine multiple objective linear programming, and data envelopment analysis. In multiple objective linear programming the purpose is to help a decision maker find the most preferred solution on the efficient frontier. Especially, this book provides an introduction to incorporating preference information in data envelopment analysis with a special emphasis in value efficiency analysis. In the beginning, the authors explain the motivation and introduce the basic concepts, the basic ideas and models of data envelopment analysis. Moreover, they give ideas to extend those models to deal with the preference information of the decision maker. The efficient frontier and production possibility set concepts which playing an important role in all considerations are presented. The approaches introduced in this book are inspired by multiple objective linear programming. The authors presented comparison and contrast data envelopment analysis and multiple objective linear programming, providing some cornerstones for approaches presented later in the book. Last chapters of this book discuss the traditional approaches to take into account preference information in data envelopment analysis. Value efficiency is introduced, and discusses practical aspects. Moreover some extensions are presented. Especially, value efficiency is extended to cover the case when a production possibility set is not convex. In addition to theoretical considerations, numerous illustrative examples are included. The applications included in the book present the suitability of the model in real-world decision making. The first one considering efficiency of academic research, the second one discusses benchmarking and applies value efficiency analysis into efficiency of parishes, and the third one investigates the efficiency of bank branches using non-convex value efficiency analysis.