Extension on Regularity Condition in DEA Models

Data envelopment analysis (dea) is a method to estimate the relative efficiency of decisionmaking units(dmus) performing similar tasks in a production system that consumesmultiple inputs to produce multiple outputs. so far, a number of dea models have beendeveloped: the ccr model and the bcc model are well known as basic dea models. thegeneric regularity condition for the ccr model is introduced so that each weakly efficientdecision making unit in the ccr model of data envelopment analysis is also ccr-efficient.in this study, we give a general regularity condition under which each weakly efficient decisionmaking unit in the bcc model and additive model of data envelopment analysis isalso bcc-efficient and additive-efficient,respectively.