Research On Rail Passenger Flow Allocation And Segmentation Model Based On IC Card And Mobile Payment Data

With the continuous development of social economy and the continuous improvement of the level of urbanization,the rail transit network has formed gradually,the opportunities for rail passengers to travel and transfer have increased,and the number of routes to choose from has increased,complicating the temporal and spatial distribution of rail transit passenger flow,even causing extreme congestion in some areas.Increasing the pressure of rail transit operation management and operational safety risks,it is difficult for the traditional passenger flow distribution model to describe the travel behavior of passengers accurately under the influence of multiple factors such as transfer and congestion.IC card and mobile payment data is used to research on passenger travel routes and passenger flow distribution,and analyzes the travel behavior of rail transit passengers in this paper,which is of great significance for improving the efficiency of rail transit operation organization and management.First,based on graph theory and complex network theory,the Space L method is used to analyze the topological structure characteristics of the rail transit network in this paper,and the rules and search methods for effective paths is determined;by preprocessing the rail IC card and mobile payment data,the distribution characteristics of passenger flow in two dimensions of time and space are obtained,the time characteristic is that each site has different attributes in the daily range,the morning peak and evening peak forms are different.The intensity gradually decreases from the city center to the outer periphery,and the tide phenomenon in the surrounding areas is obvious;secondly,from the internal factors such as the age and income of passengers and the external rail transit operation time,congestion and other factors,the travel behavior of passengers is analyzed to extract the number of travel days,average travel time,travel OD coverage as clustering indicators,the GMM clustering algorithm is used to divide passengers into four categories: commuter and non-commuter,it’s found that commuter passengers pay more attention to travel time factors,non-commuter passengers pay more attention to travel comfort factors;then,based on stochastic utility theory and factors such as transfer and congestion that affecting passenger route selection are quantified,the generalized travel cost function for passengers is established.In order to solve the problem of independence of probability and irrelevant alternatives,the path scale is added to the MNL model,according to the existing sorting methods and rules,a passenger flow distribution and sorting model based on the path scale is constructed;finally,the certain rail network in Chongqing is selected.The card swiping data on weekdays is analyzed as an example,and the passenger flow distribution ratio and clearing ratio between OD pairs are calculated,which intuitively reflects the differences in the route choices of different types of passengers.The calculation results show that the average error of the passenger flow allocation and sorting model based on the path scale is 4.18%,the average error is reduced by 32.58%and 1.52% compared with the traditional single-path allocation method and multi-path allocation method,which verifies the accuracy of the model constructed in this paper.The stochastic utility theory is applied in this paper.Based on rail IC cards and mobile payment data,a passenger flow allocation and sorted model is proposed,which could describe the travel route selection of different types of passengers,realize differentiated passenger flow allocation for different types of passengers,and improve rail passenger flow allocation.It enriches the rail passenger flow allocation method of the transportation discipline,and provides a certain decision-making theory and method for rail line optimization,operation organization management,and ticket clearing calculation.