Analysis Of Cab Route Choice Preferences Considering Travel Mode Differences

As a fundamental issue in urban traffic planning,traveler path planning is of great significance to improve the overall operational efficiency of urban traffic and alleviate the congestion of urban road networks,affecting the economy of individual travel,the rationality of road flow distribution and the scientificity of traffic management planning.Many studies have shown that travelers’ route selection behavior is not only determined by various factors such as travel cost and road attributes,but also potentially influenced by personal preferences and travel inertia.Thanks to the development of GPS technology,travel path selection behavior can be reflected on the road network realistically and accurately,and by studying the characteristics of travelers’ path selection behavior,we can provide effective technical support for path planning.The purpose of this paper is to analyze the travel path selection preferences of drivers and reveal the microscopic mechanism of travel path selection behavior based on the stochastic parametric logit model,so as to obtain the optimal path that is more in line with the reality.The main research elements include the following:Firstly,the path data and cab trajectory data obtained from the Open Street Map(OSM)platform are filtered and extracted.Based on this,the travel paths are matched to the electronic map,and the speed of each road segment is calculated for each time window.Next,the shortest distance path and the shortest time path are constructed by considering the length of the road section and travel time impedance.The overall road attribute characteristics of the three paths of the actual travel path,the shortest distance path and the shortest time path,the distribution of the actual travel path and the shortest path in terms of distance difference and time difference on different attribute classifications are analyzed,and a spectral clustering algorithm is used to classify the travel path selection pattern based on the shortest path,which is finally divided into three categories: attribute optimal,biased time distance type and others.After that,correlation,covariance and significance tests are conducted on the influencing factors,and the random parameter logit model is used as a framework to calibrate the parameters of the above travel path selection patterns and analyze the influence of different path attributes on travel patterns,and the results show that people are willing to pay a certain time distance cost to choose the road with better attributes,the relative difference in the number of turns and the relative difference in the number of nodes are negative utility for path selection patterns The relative difference in the number of turns and the relative difference in the number of nodes is a negative utility for the route choice mode,and the relative difference in distance and time is a positive utility.Finally,under the time and distance thresholds of each mode,considering the utility of each path under different travel modes,the K shortest circuit algorithm is constructed to select alternative paths,and finally the path with the greatest utility under each travel mode is selected as the recommended path.