A MULTIPLE OBJECTIVE PROGRAMMING APPROACH TO DATA ENVELOPMENT ANALYSIS

In this paper a multiple objectives programming method is applied to improve the discriminating power of classical Data Envelopment Analysis (DEA) method. Unlike the classical DEA model often producing many relatively efficient decision making units (DMUs), this new approach enjoys more discriminating power, which results in less DMUs with efficiency ratio as 1. In this approach, every DMU's efficiency evaluation is viewed as one objective function to be maximized. A set of common multipliers, input and output weights, can be located not difficultly by using the fuzzy multiple objectives programming approach. In comparison to the number of programming works being same as the number of the DMUs in traditional DEA model, the new approach just needs to solving multiple objectives programming problem once no matter how many DMUs are. Apparently, the new approach is comparatively suitable to solve a problem with a large number of DMUs.