A real-time bus arrival time information system using crowdsourced smartphone data: a novel framework and simulation experiments

This paper proposes a novel framework for developing a real-time bus arrival time information system, using crowdsourced bus information contributed by bus passengers. On the one hand, passengers can derive the real-time information via their smartphones. On the other hand, they can provide some bus data in return. Particular characteristics of the participatory-based bus data are introduced. Also, a number of data processing steps are proposed in the framework to handle the data characteristics, which pose extra difficulties in real-time bus arrival time prediction. The proposed system is evaluated using simulated bus data sets. Practicality of the system is investigated in terms of prediction accuracy based on different participation percentages of bus passengers.