Innovation efficiency evaluation of industrial technology research institute based on three-stage DEA

The industrial technology research institute (ITRI) has become an important platform to promote the transformation of scientific and technological achievements in universities. The influence of environmental factors and random errors is the main problems in the evaluation of ITRI's innovation efficiency, which greatly affects the evaluation results. At present, there are few studies on the ITRI's innovation efficiency, and most existing works do not eliminate environmental and random factors, resulting in biased results. In this paper, for the evaluation of the innovation efficiency of ITRI, the three-stage data envelopment analysis model is utilized. Firstly, in the first stage, based on the original input and output data, the efficiency of each decision making unit is measured by using the input-oriented classic DEA model. Secondly, the stochastic frontier model is constructed in the second stage to eliminate environmental factors and random errors in the original input and output data. Finally, in the third stage, the adjusted input data is brought into the calculation of the first stage to obtain the real innovation efficiency value of ITRI. The extensive experiments are constructed on the relevant indicator data of 80 Chines ITRI, whose results indicate that the evaluation of innovation efficiency is indeed affected by environmental factors and random errors. After eliminating the influences of environmental factors and random error, the innovation efficiency shows a decreasing tendency, which is more in line with Chinese national conditions.