基于AHP和ANN的不同车型收费系数确定

不同车型收费系数反映了不同车型用户承担收费公路通行费的比价关系,合理的收费系数,应该实现责任和效益的双重公平.在建立影响不同车型收费费率的指标体系前提下,综合考虑定量和定性因素,分别建立了关于道路破坏和级差效益指标的两个BP神经网络,利用层次分析法,构造BP网络的输入和输出样本,通过训练获得稳定的网络结构和权重,然后输入不同车型代表性指标值,确定不同车型的收费系数.实验表明,该方法不仅理论上可行,而且实践上得到了不同车型收费系数的合理取值区间.

A Mixed AHP-ANN Method to Determine Toll Rate Coefficient

Different toll rate coefficient should reflect that how much toll falls on different vehicle class users in order to achieve equality of cost and profits.Taking into account of both objective and subjective factors,road cost and user profits index system is established.Then analytic Hierarchy Process and BP Neural Network are used to determine toll rate coefficients.Experiments show that this method is not only reasonable and feasible,but also has good effects.