Logistics company performance evaluation by BP neural network and DEA

In the process of evaluation logistics company performance by BP neural network, the more difficult problem is to determine the evaluating sample expectations, while the C²R DEA cross-evaluation model is able to identify the rank and grade of evaluating samples (the unit of making decision). In this paper, we set the cross-model evaluation results to the sample expectations, integrate BP neural network with cross-evaluation model and construct the performance evaluation model of the logistics company, and use 23 Chinese logistics companies in 2007 as samples to carry out an empirical study. The results show that evaluation results are consistent with the actual situation, and the comprehensive weights reflect the degree of the performance indicators´ influence basically.