分路段差异化收费条件下货车司机出行路径选择意愿模型

为解决货车司机出行路径选择行为缺少考虑差异化收费政策影响的问题，通过出行费用、出行距离、优惠折扣、收费系数和是否关注差异化收费政策这5个变量刻画差异化收费，通过构建分路段差异化收费条件下货车司机出行路径选择的RF模型。以银昆高速（G85）昭通至水富段及麻水线与昭麻二级路问卷调查数据为基础开展实证分析。结果表明：所构建的分路段差异化收费条件下货车司机出行路径选择的RF模型的分类准确率优于AdaBoost模型、GBDT模型和传统Logit模型；整体重要度方面，优惠折扣是影响货车司机出行路径选择最主要的因素(38.56%)，出行费用(13.55%)、出行距离(10.06%)及出行时段(7.08%)对意愿结果存在显著影响；同时，部分分路段差异化收费变量对货车司机出行路径选择意愿存在明显的阈值效应。

Truck driver 's travel route choice willingness model under the condition of differentiated tolls by road section

In order to solve the problem that the travel route choice behavior of truck drivers lacks consideration of the impact of differentiated charging policies, the differentiated charging is characterized by five variables : travel cost, travel distance, preferential discount, charging coefficient and whether or not to pay attention to differentiated charging policies. The RF model of truck drivers '' travel route choice under differentiated charging conditions is constructed. Based on the questionnaire survey data of Zhaotong to Shuifu section of Yinkun expressway ( G85 ) and Mashui line and Zhaoma secondary road, the empirical analysis is carried out. The results show that the classification accuracy of the RF model is better than that of AdaBoost model, GBDT model and traditional Logit model. In terms of overall importance, preferential discount is the most important factor affecting the travel route choice of truck drivers ( 38.56 % ). Travel cost ( 13.55 % ), travel distance ( 10.06 % ) and travel time ( 7.08 % ) have a significant impact on the willingness results ; at the same time, there is a significant threshold effect on the willingness of truck drivers to choose travel routes.Key words: Differential tolls by section; truck driver travel behavior; random forest model; machine learning model; nonlinear relationship