Travel trajectory detection of travelers in urban rail transit based on reference passengers

Abstract

Urban rail transit is the backbone of metropolitan transportation. With continuous growth of passenger flows and deepening of networked operations, its operation and management face new opportunities and challenges. Passenger travel trajectory is the basis for analyzing passenger flow distribution and travel behaviors, but characteristics of one-pass transfer system and non-check-in seat bring great difficulties to passenger trajectory analysis. The Automatic Fare Collection (AFC) system records massive passenger travel data, and the data mining method in the area of big data provides a new way of thinking for tracking passenger travel trajectories. By mining the passenger travel data recorded by the AFC system and combining the relationship between urban rail passenger flow and train, this paper proposes a “passenger flow-train” matching algorithm based on reference passenger. The method can be applied to the daily analysis of urban rail passenger flow, and provides support for the organization of operation management agencies.