Estimating returns to scale using non-radial dea models

Abstract

The concept of returns to scale (RTS) is defined as the ratio of the proportionate changes in outputs over the proportionate changes in inputs. By considering the following two facts the current paper develops some non-radial data envelopment analysis (DEA) models to address a new concept of RTS termed the component RTS :a) The proportionate changes in input will not necessarily cause the proportionate changes in outputs; b) If it is desired for decision maker (DM) to find out about the rate of increase in a specific component of output vector after exerting changes in inputs, the radialbased models will not be able to make this wish come true. In other words, the main objective of this work is to seek the disproportionate changes, coming in to existence in any individual components of output vector, through exerting changes on inputs of under evaluation unit. The suggested models are used in a case study that is focused on RTS estimation of some bank branches.