Risk analysis for highway project based on multiple-dimension neural network

Based on the detailed analysis of the. internal and external factors that influence highway projects, this paper builds a system about highway project risk assessment index, structures a multi-dimensional-input neural network model and its updated algorithm and establishes the rules for framework's adjusting and optimizing. Furthermore, the utilization of fuzzy integrative assessment confirms the numerical value of various risky factors, and this assessment system combined with the structured neural network model can comprehensively and effectively assess the risk probability of highway project investment, then this paper presents a previous risk-alarm system. In this paper the authors try to provide a method and framework for the investors to solve the problems about the integrative assessment of highway project risk and about avoiding the subjectivity of confirming the risky factors' weight.