Spatio-temporal trajectory estimation based on incomplete Wi-Fi probe data in urban rail transit network

Area of Science:

• Transportation Science
• Data Science
• Network Analysis

Background:

• Automatic fare collection systems provide limited passenger movement data.
• Wi-Fi probe data offers richer insights into passenger activity within transit networks.
• Estimating detailed passenger trajectories from incomplete Wi-Fi data is challenging.

Purpose of the Study:

• To develop a methodology for estimating personalized and timely passenger spatio-temporal trajectories.
• To address data deficiencies in Wi-Fi probe data for trajectory estimation.
• To leverage multiple data sources for accurate passenger movement reconstruction.

Main Methods:

• Description and defect analysis of Wi-Fi probe data.
• Development of the n-gram method for inferring missing location information.
• Design of an estimation algorithm integrating network topology, Wi-Fi data, and train schedules.

Main Results:

• Successfully estimated 93% of unique physical passenger routes in a complex urban rail network.
• Reduced the number of feasible spatio-temporal trajectories to one or two for 80% of passengers.
• Validated the methodology using both simulated and real-world data.

Conclusions:

• The proposed methodology effectively estimates passenger spatio-temporal trajectories from incomplete Wi-Fi data.
• The approach enhances the detail and accuracy of passenger movement analysis in urban rail transit.
• Identified potential applications for trajectory estimation in transit operations and planning.