Using GPS tracking data to validate route choice in OD trips within dense urban networks


There are presently several companies that provide processed or raw global positioning system (GPS) measurements generated by fleets of commercial vehicles, internet applications or automobile companies. The aim of this paper is to deepen our understanding of GPS data applicability in transportation modelling by providing systematic and quantified insights into the representativeness of collected data in describing individual’ route choices. Unfortunately, real data often contain noise, uncertainty, errors, redundancies or even irrelevant information. Useless models will be obtained when built over incorrect or incomplete data. This is why pre-processing is one of the most critical steps in data analysis. Yet, pre-processing has not been properly systematized, which is why this paper focuses on the necessary steps for pre-processing GPS tracking data together and puts forth a proposal for systematizing it. Furthermore, the aggregation level is at waypoint location for low latency GPS positions along the trip trajectory. Travel time reliability on OD paths is addressed, along with other OD path characteristics. Dense urban networks pose multiple possibilities for route choice, and data from new technologies offers an opportunity to understand route choice behaviour.

Transportation Research Procedia