【摘要】 采用系统捕捉和随机捕捉两类捕捉形式对出行方式的时间感知差异进行建模分析,以验证出行方式选择中时间认识的个体差异性,为更准确的交通方式划分预测提供模型参考。系统捕捉通过在Logit模型中分别设置性别、收入与出行时间的交叉变量以捕捉性别和收入不同导致时间价值认识的差异;随机捕捉通过假定出行时间参数分别服从正态分布、均匀分布、对数正态分布以及约翰逊分布4种随机分布形式,利用Mixed Logit模型捕捉出行者对时间感知的差异。结果表明:出行方式选择中出行者对时间重要性认识存在差异,男性比女性对时间要求高,高收入者比低收入者对时间估值更高;随机捕捉较系统捕捉更能捕捉到出行时间的感知差异,且约翰逊分布较其他分布对时间感知差异性的模拟更优。更多还原

【Abstract】 Two kinds of model specifications were established to model the travel mode choice behavior to detect the presence of individual preference heterogeneity.The first approach was to specify the model with each attribute parameter as a function of observed socio-economic characteristics of the individual,which captured the systematic taste variations.And the function form were set as Gender×Time and Income×Time separately.The second way was to capture the random taste variations through the specification of a Mixed Logit model,in which four distribution were considered for time parameter: normal,uniform,log normal and Johnson.The results show that: individual time preference heterogeneity indeed exits;male would pay more attention to travel time than female when facing the choice of the travel mode;those with higher income value the travel time much more than those with lower-income;random taste variation capturation results are better than the systematic ones and Johnson distribution is the best to capture the random taste variations comparing with other distributions.更多还原