01663591

Component

Automatic Fare Collection (AFC) systems record the time and location information when a passenger enters or leaves the urban rail transit system by swiping his/her smart card. The AFC transaction data records passengers’ trip information exactly and can be used for many researches. This paper aims to estimate the most likely space-time path based on AFC transaction data. A complete passenger’s travel path is consisted of four components: access walk, in-vehicle, egress walk and transfer walk. By constructing a time-extended network based on train timetable data, a space-time path model is formulated to simulate passenger’s trajectory, which tells a passenger’s movements among activity locations with respect to time. Then, a time-dependent maximum likelihood space-time estimation model is proposed to estimate the most likely space-time path for all passengers. Considering the computational efficiency and the characteristic of space-time path, the authors propose an improved Dijkstra algorithm to solve the space-time path estimation problem. Real-world AFC transaction data and train timetable data from Xi’an subway is used to verify the proposed model and algorithm.

This paper was sponsored by TRB committee AP065 Standing Committee on Rail Transit Systems.

18-01144

English

17p

2018

Transportation Research Board 97th Annual Meeting

Digital/other

Figures; Maps; References; Tables

Automatic fare collection; Data mining; Passenger traffic; Subways; Trajectory

Xi'an (China)

Data and Information Technology; Operations and Traffic Management; Passenger Transportation; Railroads

Transportation Research Board Annual Meeting 2018 Paper #18-01144

TRIS, TRB, ATRI

Jan 8 2018 10:17AM