Research On The Stickiness And Preference Of Public Transport Passengers’ Travel Route Choice By Fusion Of IC Card And GPS Trajectory Data

Intensive travel mode caters to the future development path of China’s large-scale densely populated cities,and high-capacity public transport can solve the practical problems of large population and compact land use.Travel is a basic human need,and passengers choose appropriate routes according to their travel utility.With the development of intelligent urban transportation,the use of public transport travel data to reproduce the decision-making behaviour of structured travel individuals,combined with data to fully explore the factors influencing public transport passengers’ route choice,travel stickiness and individual preferences,is an important topic for public transport system research.The thesis first correlates spatio-temporal constraints based on the IC card of the public transport system,vehicle GPS trajectory data and urban bus line network data,completes the projection of passenger boarding and alighting,transfer station information,fuses to extract passenger travel OD,and extracts bus passenger travel routes based on bus vehicle trajectories between OD stations.Subsequently,passenger travel characteristics and attributes are studied,and passenger route choices are analysed according to the traveller’s own attributes,travel attributes and route attributes.The origin and destination Euclidean distances,travel frequencies and journey times are selected to initially analyse the proportion of interchange trips for different types of passengers in order to compare the differences in passenger choice of interchange routes,and the comparison reveals that passengers’ actual travel routes are not the theoretical shortest travel routes.Furthermore,from the perspective of individual preferences showing habitual behaviour,the concept of stickiness index is introduced to quantify the results of passengers who always choose the same route to travel(high stickiness)versus those who choose a more diverse route pattern(low stickiness),and to measure the habitual behaviour of passengers in choosing routes between ODs.Finally,a familiarity function is introduced as the familiarity factor of passengers with the road network,and the MNL model of passenger route choice is constructed by combining the route attributes,calibrating the route attribute parameters,and using the sensitivity coefficients of different groups of people and travel distances to portray passenger travel preferences.At the same time,this paper constructs differentiated service scenarios based on the route attribute options and compares passengers’ choice behaviour when faced with different travel route attribute options.This paper reduces passengers’ travel routes from massive public transport data,analyses the influence of travel attributes on passengers’ travel route choices,portrays passengers’ route choices from macro habitual for and micro individual preferences,and provides an idea for studying individual route choice exploration under large-scale data conditions.