Context-aware multi-modal route selection service for urban computing scenarios

Abstract

The rapid urban population growth introduces novel challenges to urban mobility scenarios, requiring innovative and connected solutions to relieve traffic congestion and enhance transportation efficiency, using unusual roads and public transportation modals. For instance, multi-modal routes allow more flexible modal combinations to offer smart multi-modal mobility. In this context, the route selection service must rely on different context information, such as criminality, accidents, air quality, and others, where Internet of Things technologies introduce active and ubiquitous sensing of many contexts along the urban scenario, providing data for statistical analysis to offer a safer and healthier urban reality. However, designing an efficient multi-modal route selection service that considers different context information to offer personalized routes is important. In this article, we describe a context-aware route selection service that considers adequate contextual information to provide routes according to the user’s preference. The multi-modal route selection service applies a multi-criteria method to give different degrees of importance to each criterion based on the user profile (i.e., Worker, Green, Safe, and Tourist). We present extensive evaluation results from applying our multi-modal route selection approach to a London dataset. The approach successfully enables user-preferred transportation choices using four balanced selection profiles and ten route features. The proposed multi-modal route selection service demonstrates better performance in terms of economic, ecological, and time-saving metrics for each user profile compared to a greedy selection manner.