Efficiency Improvement through Distance Friction Minimization in Data Envelopment Analysis -an Application to the Tourism Sector in Italy-

Data Envelopment Analysis (DEA) has become an established approach in efficiency improving strategies in industrial and regional programming. The aim of the present paper is to propose a new and recently developed Distance Friction Minimization (DFM) model in the context of DEA in order to provide an appropriate efficiency-improving projection model. The standard DEA models, including the Charnes Cooper Rhodes (CCR) model and the Banker Charnes Cooper (BCC) model, have thus far only focused on a uniform input reduction or a uniform output augmentation function in the improvement projections. In our approach, the DFM model is developed as a more flexible tool that offers a new contribution to efficiency enhancement by deploying a weighted projection function, while it may address both input reduction and output increase. In our study a generalized distance friction will be presented to assist a Decision-Making Unit (DMU) in improving its efficiency performance by a proper movement towards the efficiency frontier surface. The direction of efficiency improvement depends on the input/output data characteristics of the DMU. Each of these characteristics may have a different weight for the DMU, so that he is essentially confronted with a Multi-Criteria Analysis (MCA) problem. To achieve an appropriate rise in efficiency, it is thus necessary to take into account the most appropriate input/output weights of these characteristics. The paper then sets out define the projection functions for the minimization of distance frictions, using a Euclidean distance in weighted spaces. An appropriate form of multidimensional projection functions serving to improve efficiency is given by a Multiple Objective Quadratic Programming (MOQP) model which is a proper method to minimize the aggregated input reduction frictions and or the aggregated output augmentation frictions. Another new element in the paper addresses the so-called fixed factor phenomenon, where some inputs cannot be flexibly manipulated by a DMU. The above-mentioned extended DEA model approaches will be empirically illustrated by using a data set on Italian tourist destination regions where the aim is to increase the efficiency of tourist strategies in these areas on the basis of a comparative study on the performance -in terms of both input and output indicatorsof these regions. Various numerical results are presented and interpreted.