Understanding driver deviations from intended routes under social routing recommendations in navigation apps

Abstract

To understand drivers’ route choice behavior under social routing recommendations, this study investigates data from a mobile navigation app that captures drivers’ stated route intentions before trips and their actual trajectories during trips. An inconsistency can occur when drivers state they will follow a specific route but then deviate from it, creating an intention-behavior gap. This study analyses real-world data from experiments in Helsinki, Finland, and Amsterdam, the Netherlands. As travelers may respond differently to social routing depending on their individual orientation, this study focuses on one specific implementation designed to improve livability and safety. By comparing the stated intentions with actual behavior at the trip level, this study provides empirical evidence on the intention-behavior gap, largely unexplored in transportation research. We define several levels of deviation and use random-effects binary logit models to examine factors associated with deviation. The findings underscore the role of travel time differences between the social and fastest routes, as well as trip contexts such as rush-hour departure time and trip purpose, providing insights for designing interventions to improve compliance with social routing recommendations.