A Performance Evaluation On Coal Firms In China Based On Context-dependent DEA Approach And This Approach Improvement

This paper does some researches on the context-dependent DEA approach, including operating efficiency analysis of coal firms based on context-dependent DEA approach and an improved approach of context-dependent DEA under variant returns to scale. In our application research, we choose coal industry to evaluate operating efficiency since coal firms are facing a serious problem of scant demand and excess supply in the development with Chinese economy stepping into the period of new normal. Our aim is to provide useful operating efficiency information and valuable analysis for managers and policy makers in order to solve the problem of excess capacity. In this research,24 coal firms are chosen as evaluation sample and we build evaluation indexes, collect related operating data in nearest three years and measure the attractiveness and progress of some companies based on context-dependent DEA approach. Through our analysis and discussion, meaningful and useful findings are summarized including that which strategy is fit to adopt for a specific coal firm: keeping current status, reforming and upgrading business or eliminating and integrating. Compared with previous related researches, our data is newer and traditional DEA models are abandoned. We believe that our work is important and valuable for coal companies, managers and policy makers. On research of improving context-dependent DEA approach, we point out two problems of measuring attractiveness and progress when directly extending original models into variant returns to scale versions. One is that in some situations, it may not be possible to determine the attractiveness of a DMU that logically must be attractive in that context. The other problem is that the progress measure may not help an inefficient DMU project to a Pareto-efficient frontier. A small numerical example is used to illustrate these two problems. In order to overcome these deficiencies, the concept of closest target is introduced in our models. Besides, we extend the definition of the attractiveness and progress so as to improve DMUs’performance with less effort. We also compared our approach with original models and slacks-based measure ones in this paper to show the merits of our work. A classic application of printers is reused to show the fitness of our models. Generally speaking, our researches are meaningful and valuable on practice and theory areas both.