Abstract: Mixed traffic flow, consisting of cars, trucks and other like, is one of the major characteristics of China’s road traffic. The complex composition and significant disturbance have received extensive attention from researchers. The paper develops a multi-class network equilibrium model. In which, travel time and road prices are separately given for different classes of users, and there is no interaction among them. The volume delay functions based on the variable PCE of mixed traffic are nonlinear, non-smooth, and asymmetric. The problem is formulated as a nonlinear variational inequality model, and the algorithm-the method of successive averages (MSA), is used to solve the problem. Finally the algorithm is implemented in the EMME/3 transportation planning software, and applied to the actual highway network, and the conversion coefficient of linear and nonlinear the transport assignment results are compared. Studies have shown that traffic assignment with non-linear conversion coefficient is more realistic of the traffic flow distribution.

Key words: traffic engineering, network equilibrium, multi-class traffic assignment, algorithm, variational inequality problem

摘要： 我国多数道路是典型的混合交通，交通组成复杂，相互间干扰现象严重. 含有小汽车、货车和其他车型的混合交通备受研究者的关注. 本文构建一个考虑道路收费的多用户网络均衡模型，在模型中，各类用户拥有各自的时间延误函数和道路收费，且相互影响不同. 折算系数（PCE）为可变的，构建的道路延误函数是非线性，非连续和非对称的. 混合交通网络均衡模型采用变分不等式构模，用相继平均（MSA）方法进行求解. 最后，在交通规划软件EMME/3中实现该算法并应用到实际公路交通网络，并对线性折算系数和非线性折算系数下的交通网络分配结果进行了对比. 研究表明，非线性折算系数更加符合实际区域网络交通流分布.

关键词: 交通工程, 网络均衡, 多用户交通分配, 算法, 变分不等式问题