Exploring potential car trips for long-distance school escorting using smart card data and a household travel survey

Abstract

Encouraging students to commute by the metro can effectively reduce household car use caused by long-distance commuting to school. This article focuses on the frequency of metro use by groups of students commuting to school based on the assumption that students who use the metro may occasionally be driven to school by their parents. For the 1st time, we propose a school metro commuter identification process that considers the potential behaviour of escorted students, and we study the potential car trips for long-distance school escorting in Nanjing (China) using Smart Card Data (SCD) and a household travel survey from Nanjing. 3 clusters of students who use the metro for commutes to school are identified by frequency of use for possible escorting behaviour based on the commuting day. As possible factors influencing the 3 frequency groups, usage pattern of the metro, entry time, travel duration and the school–housing relationship are extracted from SCD. Furthermore, a multinomial logistic regression model is used to examine the significant factors that influence the grouping of students. The results show that students who use the metro occasionally for a long commuting distance to school are more likely to be escorted to and from school by their parents, especially to school. The later the entry time is to the metro, the more likely that students are to be escorted to school. Additionally, a long school–housing travel duration/distance significantly contributes to parents’ car trips for commuting. The results of this article are valuable for transport policy to reduce car use for long-distance school trips.