A Range Adjusted Measure of Super-Efficiency in Integer-Valued Data Envelopment Analysis with Undesirable Outputs

Abstract DEA (data envelopment analysis) models can be divided into two groups: Radial and non-radial DEA, the latter has higher discriminatory power than former. The range adjusted measure (RAM) is an effective widely used approach. However, to best of our knowledge, there no literature on integer-valued super-efficiency RAM-DEA model, especially when undesirable outputs are included. We first propose model with then extend this outputs. Compared other models, novel have many advantages: 1) They non-oriented which enable decision makers simultaneously non-proportionally improve inputs outputs; 2) handle variables outputs, so results obtained more reliable; 3) easily as it based linear programming; 4) accurately rank efficient DMUs. proposed applied evaluate efficiency China’s regional transportation systems (RTSs) considering number transport accidents (an output). help performance inefficient RTSs analyze strengths RTSs.