轨道交通乘客个性化出行路径规划算法

Personalized Route Planning Algorithm for Urban Rail Transit Passengers

随着城市轨道交通网络的不断完善,可供乘客选择的轨道交通出行路径日益增加,乘客出行路径决策愈加复杂.本文在分析轨道交通服务水平变量对不同属性乘客出行路径选择行为影响的基础上,提出轨道交通乘客个性化出行路径规划算法.首先,基于非集计理论构建针对不同类别乘客的路径选择模型,该模型综合考虑乘车时间、换乘时间、换乘次数、车内拥挤度及个人属性等因素对乘客路径选择行为的影响.其次,基于不同类别乘客的路径选择行为差异,构建考虑车内拥挤度变化的乘客个性化出行路径动态规划算法,为不同属性乘客规划广义出行时间最小的路径.最后,基于广州地铁数据对算法进行验证.结果表明,该算法针对乘客个人属性规划的最优出行路径,更加贴合乘客的出行心理.

With the rapid development of urban rail transit network, increased available routes make passengers＇ trip decision-making become more and more difficult. This study proposes a dynamic metro route planning algorithm with the least generalized cost, in which network LOS variables and personal characteristics are taken into consideration. Firstly, based on the disaggregate choice theory, the metro route choice models for different types of passengers are established with the consideration of LOS variables（e.g.in-vehicle travel time, transfer time, number of transfers, in-vehicle passenger density, etc.） and personal characteristics（e.g. age, trip purpose, etc.）. Then, based on the proposed models and the time-varied sectional flow volume, a dynamic personalized route planning algorithm is proposed, which is expected to generate the optimal route with the least generalized time for each type of passengers. Finally, the proposed algorithm is evaluated in Guangzhou Metro conditions. The results indicate that the algorithm can provide passengers with the more reasonable route corresponding with their characteristics, and expresses passenger＇s route choice preference more precisely.