Applying an Improved BP Network to Risk Assessment of Performance for Expressway Management Corporations

In the process of operating an expressway project for a management corporation, there are a large number of uncertainty factors influencing the management performance. Based on the detailed analysis of the internal and external factors and the influencing route, an index system of risk assessment of performance was built for expressway management corporations. The numerical value of various risky factors can be confirmed by utilizing fuzzy integrative assessment. Furthermore, a neural network model was designed and its algorithm was improved to be suitable to multi-dimension inputs in this study. The evaluation system based on the model can comprehensively and effectively assess the probability of management risk, and then an anticipated risk-alarm system was developed. As an example, the model was used to assess the risks of performance for expressway management corporations, and the evaluation was comparatively objective. The model can effectively avoid the subjectivity of confirming the risky factors´ weight. A new method, which can effectively solve the problem about the integrative assessment of management risks, was proposed for the managers and superintendents of expressway management corporations.