Inter-bus-station operation time interval prediction method based on support vector machine

Abstract

The invention discloses an inter-bus-station operation time interval prediction method based on a support vector machine. Firstly data cleaning is performed on bus GPS original data; then bus arrivaltime is extracted from the data and the inter-bus-station operation time of the bus is calculated; the relevant information is selected to establish an inter-bus-station operation time interval prediction model input data set; two support vector regression machines are established to predict the upper and lower bounds of the bus operation time; parameter optimization is performed on the support vector machines by using a particle swarm algorithm, and the higher effective coverage of the prediction interval and the lower average width of the standard prediction interval act as the parameter optimization objectives; and the final inter-bus-station operation time interval prediction model is constructed according to the optimal parameters obtained by the PSO algorithm. Real-time and accuratebus arrival time interval prediction can be provided for the travelers under the uncertain situation so that planning and selection of the traveling route can be facilitated for the travelers.

Description

technical field [0001] The invention belongs to the field of public traffic management optimization, in particular to a method for predicting the running time interval between bus stops based on a support vector machine. Background technique [0002] Alleviating urban traffic problems by giving priority to the development of public transport has become a long-term development strategy of urban transport in my country. Bus travel has the advantages of low cost, large capacity, and low pollution. It can efficiently use the city's road information resources, effectively alleviate the traffic congestion on urban roads, and reduce environmental pollution. However, in our country, bus delays occur frequently, especially in morning and evening peak hours, which seriously affect the punctuality of bus arrival time and make residents unwilling to choose bus travel. Therefore, real-time and accurate prediction of the arrival time of buses can not only facilitate the planning and sele...

AI Technical Summary

Problems solved by technology

[0004] Purpose of the invention: In order to solve the problem of insufficient accuracy of bus arrival time estimation in the above-mentioned prior art, the present invention provides a method for predicting the running time interval between bus stops based on support vector machines. The method uses the bus GPS data to predict the interval The effective coverage rate and the average width of the interval are taken as the optimization objectives, and the uncertain factors in the running time between bus stops are considered to establish an interval prediction model based on support vector machines

Images