Modeling Social Network Influence on Joint Trip Frequency for Regular Activity Travel Decisions

Abstract

Interest in activity-based travel demand modeling has recently increased significantly because of the level of accuracy offered by this type of modeling and its applicability in travel behavior research. Understanding the linkage between social influence and travel behavior enables efficient characterization of discretionary activities that account for a major fraction of total urban trips. In particular, some recent studies looked at the social network structure of individuals and measured the influence that social network members have on the performance of social activities. However, it is expected that the process of taking joint trips (trips that individuals take with their network members) has an intrinsic social context for different activities in general (not only social activities). In this regard, this study used an egocentric (i.e., personal) network approach to explore empirically the influence of personal network characteristics on the frequency of weekly trips that individuals (egos) took part in with their personal network members (alters). With the help of zero-inflated Poisson models and egocentric social network data, the results of this study present a framework for predicting the number of joint trips for six types of activities: work, eating out, shopping, recreation, study, and extra curricular activities. Estimation findings suggest that personal network measures such as network density, homophily, heterogeneity, and ego–alter tie attributes have a significant impact on the joint trip-making process, that is, the number of weekly shared trips in which an individual participates. The findings of this study should help practitioners implement targeted policies such as car sharing for various user groups.