Frontier models and preference elicitation in the productivity and efficiency analysis

Abstract

This doctoral dissertation aims to explore the state of the art in the field of Productivity and Efficiency Analysis and to propose an efficiency elicitation for choosing a viable direction of which outputs should be expanded and/or entries are contracted according to a predefined decision maker preference structure. The state of the art in the Productivity and Efficiency Analysis is investigated through bibliometric networks in three instances (chapters) composing this doctoral thesis: The state of the art in the Data Envelopment Analysis since the seminal paper of Charnes et al. (1978), exploring the theoretical developments, models and applications in the field; a survey on the available computational developments, software and packages dedicated to the efficiency and performance evaluation, and a meta-surveys on the current surveys of empirical application. Through an elicitation process of the decision makers preferences, multi-criteria outranking procedures with the imposition of preference, indifference and veto threshold are used to model the policy maker evaluation into directions that represent a measure of global importance in the usage of resources (instead of trade-offs). Some of the main contributions of the proposed model are the decomposition of a relative measure of inefficiency for each resource and for each outcome based on a predefined preference structure; the optimal level of contraction in the input distance function or expansion in the output distance function case; an absolute measure of inefficiency with the slacks based on a predefined preference structure; the benchmark units serving as the reference for the linear combination from which efficient practices might be inferred; and a ranking of (efficient and effective) alternatives taking into consideration the preference and pre-defined policies from the decision maker.