Urban rail transit passenger travel time chain identification system

Abstract

An urban rail transit passenger travel time chain identification system extracts passenger travel origin and destination points from transaction data recorded by an automatic fare collection system, performs effective path search, and matches all possible travel time chains of each path for each effective path in combination with train timetable information. A transaction record which is only matched with one travel time chain is selected as a reference record, the reference record is classified according to the difference between a terminal station gate and an exit gate, and the get-off exittime difference probability distribution condition of each gate is fitted. For the transaction records matched with the multiple travel time chains, the probability is calculated through a corresponding gate opening probability density function according to the selected route, the gate opening and the get-off exit time difference of each time chain, and the travel time chain with the maximum probability is selected as the travel time chain of the passenger. A complete travel time chain of passengers can be obtained, and a basis can be provided for operation management and ticket clearing of rail transit.

Description

technical field [0001] The invention belongs to the field of passenger flow simulation of urban rail transit lines and networks, in particular to an identification system for passenger travel time chains of urban rail transit. Background technique [0002] With the continuous development of my country's rail transit, rail transit in many cities has entered the stage of network operation. In order to understand the state of the road network, arrange traffic organization reasonably, and carry out accurate classification, it is necessary to be able to obtain the complete travel time chain of passengers. However, existing The current AFC system can only record the time of swiping the passenger's card when entering and exiting the station, but cannot record the route selection and train number of the passenger. In order to achieve this goal, a road network simulation model can be constructed to calculate the travel time chain of passengers. Regarding the construction of the road ...

AI Technical Summary

Problems solved by technology

Regarding the construction of the road network model, the currently commonly used method is to describe the rail transit network in the form of a graph to search for an effective path, such as a series of shortest path solving methods such as Dijkstra algorithm, FLOYD algorithm, and A* algorithm, as well as the K shortest path based on these methods. However, these methods generally have the problem of low search efficiency; for the acquisition of passenger travel time chain, most of the domestic and foreign researches focus on constructing passenger's route selection behavior with stochastic utility theory, and through research to influence traveler's route choice Factors, construct a utility function according to the influencing factors, and then use the aggregate model to calculate the probability of each path being selected

This method needs to calibrate a large number of model parameters, which requires high cost and cannot obtain the route selection of a single passenger and its complete travel time chain

Images