A Non-Radial Input Relaxation Model Approach for Evaluating Input Congestion in Data Envelopment Analysis

This paper develops a non-radial model, unified additive model approach, for evaluating decision making units in data envelopment analysis. Based on the proposed additive input relaxation model, two approaches, a two model approach and a one model approach are provided for determining input congestion. It is shown that the proposed two model approach determines the amount of congestion and, simultaneously, identifies factors responsible for congestion and distinguishes congestion amounts from other components of inefficiency. These amounts are all obtainable from non-zero slacks in a slightly altered version of the additive input relaxation model which we further extend and modify to obtain additional details. One model approach, also, determines input congestion, and is preferred from computational point of view. Numerical and empirical examples are used to illustrate the proposed non-radial methods.