Abstract: To improve the comprehensive capability of port-logistics is a hot and difficult question.So a combined method of entropy theory and BP neural was applied to research the port-logistics.The theory of entropy weight was adopted to determine the training samples and expected output,and the comprehensive capability of port-logistics were obtained based on the trained BP neural network.Instance including 7 years’ logistic data of Hekou port confirms the effectiveness and practical application value of the methods.

Key words: Port-logistics,Entropy weight,BP neural network,Comprehensive capability