Cab Empty Route Optimization Based On Traffic Big Data

With the continuous development of big data,the way people travel has been changed by the creation of intelligent transportation technology.As an important tool for urban passenger transportation,cabs provide comfortable and efficient transportation services for passengers.However,the information asymmetry between passengers and cab drivers leads to drivers blindly looking for passengers and passengers having no taxi to get,which makes the low passenger rate and high empty rate of cabs,resulting in a serious waste of transportation resources.Therefore,it is important to use the driving experience of cab drivers to obtain better paths to improve cab driving efficiency and increase drivers’ income.This thesis analyzes cab drivers’ driving experience based on cab drivers’ income and passenger hotspots,summarizes the driving experience of high-income cab drivers,provides driving strategies for low-income cab drivers,and proposes a path planning problem to reduce cab empty distance.The Floyd algorithm and Dijkstra algorithm are analyzed,and the Dijkstra algorithm with better time complexity is combined with the ant colony algorithm to solve the problem that the ant colony algorithm is easy to fall into local optimum.And the parameters of the improved ant colony algorithm and the segmentation method are optimized to obtain shorter unloaded distance.The specific work of this thesis is as follows:(1)Pre-processing of cab trajectory data to obtain cab origin Origin-Destination(OD)data.Firstly,the OD data is used to analyze the cab operation and get the hot time of passenger carrying.The spatial distribution of OD data is analyzed by rasterizing the geographic information of Shenzhen.Secondly,visualization operations are performed to obtain a view of the cab hotspot areas.Finally,the road network information is analyzed by using open map to obtain the road network topology map,which provides data basis for subsequent analysis of cab drivers’ income and summarizing driving experience.(2)Analyze cab drivers’ driving experience based on OD data to provide driving strategies for low-income drivers.Firstly,the distance traveled by each order of cab is explored,and the income of drivers is calculated and classified according to the cab fare standard.Secondly,the driving experience of drivers in passenger-carrying hours is analyzed by time and income,and the experience advantages of high-income drivers are summarized to provide driving strategies for low-income drivers.Finally,the path planning problem to reduce the empty distance of cabs is proposed.(3)The OD data ranking of the empty sections of low-income cab drivers is performed,and the hot sections are used for path planning.For the problem that the ant colony algorithm is easy to fall into local optimum,Floyd algorithm and Dijkstra algorithm are used for global optimum processing;secondly,by comparing the time complexity of Floyd algorithm and Dijkstra algorithm under large data conditions,Dijkstra algorithm with lower complexity is mixed with ant colony algorithm,and the parameters of the mixed algorithm are optimized.Finally,the segmentation method of the hybrid algorithm was improved to reduce the percentage of empty distance of cabs from 77.08% to 76.08%,and the hybrid algorithm was used in other hotspot road sections for validation,and the results showed that the percentage of empty distance was reduced from 60.77% to 59.36%.The experiment proves that the improved ant colony algorithm can shorten the empty distance and optimize the route.