Abstract: Due to the unbalance of social and economic development and the differences of regional features and statistical data quality, the utilization evaluation of power grid equipment lacks homogeneous base, and the fairness of the evaluation results is affected. A three-stage approach for utilization evaluation of power grid equipment is put forward. The approach first applies the data envelopment analysis model which has variable returns to scale in order to solve the input variable slacks. Second, the environment variables are used to fit the slacks with regression for adjusting the input variables. The adjusted input variables are substituted into the data envelopment analysis model to achieve the evaluation at last. It is demonstrated that 1) the traditional data envelopment analysis does not consider the environmental factors, which leads to the underestimation of the power grid equipment utilization in underdeveloped regions; 2) environmental factors have significant influences, which are reflected in the power grid equipment utilization; 3) the three-stage method can evaluate the power grid equipment utilization more truly and fairly.

Key words: power grid, equipment utilization, regional differences, data envelopment analysis, regression analysis