The Research On The TPL Enterprises Competitive Ability Evaluation Based On BP Neural Network

With the continuous development of logistics technology, third party logistics is arousing great attention, as a way of improved material flow rate, saving storage costs and cost of capital in transit. Scientific and comprehensive analysis and evaluation of the competitiveness of third party logistics, has become an urgent need to address the issue for the logistics enterprises.However, the discussion of the competitiveness of third-party logistics enterprises is relatively rare at present, because of the problem of the competitiveness of logistics enterprises including a large amount of uncertainty and ambiguity, we need to follow the systematic ideas and methods from all angles to evaluate and predict the problem. Yet the president general evaluation and prediction method often has been subjective and arbitrary. Therefore, BP neural network is introduced for evaluation of TPL enterprises competitive ability. The main work includes:1. Based on the analysis of third party logistics development, discuss the necessity and importance of the study about the evaluation of TPL enterprises competitive ability. The paper analyzed the advantages and shortcomings in the field, by reviewing the president research.2. According to the current definition and characteristics of TPL enterprises competitive ability in our country and through principles of evaluation index selection and evaluation index system designation principles, the index system of TPL enterprises competitiveness evaluation has been designed.3. Combined with evaluation system to create a model of TPL enterprises competitiveness evaluation. With the comparison of other evaluation method lead to BP neural network algorithm. According to TPL enterprises competitiveness index system, the paper has designed BP neural network model and gave practical evaluation procedures. In the calculation method, the neural network toolbox with MATIAB (NNT) has been used to design and calculate the network. The error of the model achieved to the desired range, by the training and testing of the studying samples.4. Using an example to validate that the indicator system and model are feasible and effective. The paper has evaluated the competitive ability of three TPL listed companies which has competitive relationship and obtained their competitiveness ranking, through the well trained BP neural network. It has been proved that the result was correct by analyzing the actual state of the three TPL listed companies.In summary, the application of BP neural network model for third-party logistics evaluation of the competitiveness of enterprises is effective and feasible, which provides a new way of thinking about the evaluation of third-party logistics enterprises competitive ability.