A Generalized Goals-achievement Model in Data Envelopment Analysis

Data Envelopment Analysis (DEA) has become an established tool in comparative analyses of efficiency strategies in both the public and the private sector. The aim of this paper is to present and apply a newly developed, adjusted DEA model – emerging from a blend of a Distance Friction Minimization (DFM) and a Goals Achievement (GA) approach on the basis of the Charnes-Cooper-Rhodes (CCR) method – in order to generate a more satisfactory efficiency-improving projection model in conventional DEA. Our DFM model is based on a generalized Euclidean distance minimization and serves to assist a Decision Making Unit (DMU) in improving its performance by the most appropriate movement towards the efficiency frontier surface. Our DFM approach aims to generate a new contribution to efficiency enhancement strategies by deploying a weighted projection function. In addition, it may address both input reduction and output increase as a strategy of a DMU. The GA model can compute the input reduction value or the output increase value in order to achieve a pre-specified goal value for the efficiency improvement in an optimal way. The above-mentioned DFM-GA model is illustrated empirically by using a data set of efficiency indicators for cities in Hokkaido prefecture in Japan, where the aim is to increase the efficiency of local government finance mechanisms in these cities, based on various input and output performance characteristics.