交通方式选择的非集计模型及其应用

应用非集计模型理论与北京居民出行调查数据,建立了交通方式选择MNL模型,并用Matlab优化工具箱中的无约束最优化函数对模型的参数进行了标定。根据所建模型计算出公交车和小汽车的时间价值,求得包含出行时间和实际金钱费用的广义费用,建立了公交车和小汽车出行选择的二肢Logit模型,分析了不同收费价格下两种交通方式所分担的交通出行比率及道路交通饱和度的变化情况,评价拥挤收费政策对优化城市交通结构,缓解城市交通拥挤的作用。研究表明,非集计建模方法能够较全面地考虑居民出行选择的各方面影响因素,尤其是将出行者的个人特性影响因素引入模型,提高了模型的预测精度和实用性。

Disaggregate modeling of traffic mode choice and its application

Based on the theory of disaggregate modeling and the residents daily travel survey data of Beijing city,a multinomial logit(MNL) model to predict the traffic mode choice was developed,and the unconstrained optimization function in Matlab optimization toolbox was used to calibrate its parameters.The time value of the bus and the car were calculated with the developed MNL model,a generalized cost including the travel time and the actual money expenses was derived,and a binary logit model was built for making travel mode choice for the bus or the car.The variation of the contributions of the two travel modes and the road traffic saturation under different pricings were analyzed,and the effects of the congestion pricing on the optimization of urban traffic structure and alleviation of the urban traffic congestion were evaluated.It was found that the disaggregate modeling approach can consider more factors affecting the residents travel mode choice,including the individual character factor,helps improve the prediction accuracy of the model.