Research on Prediction of Urban Traffic Congestion Time and Construction of Traffic System based on Survival Analysis

Abstract

With the development of economy and society, the number of cars in major cities in China has been rising sharply in recent years, and the traffic congestion problem has become increasingly serious. Aiming at the prediction of traffic congestion time, multiple linear regression and survival analysis methods are used to establish multiple linear model, Kaplan-Meier nonparametric regression model, Cox regression model and other prediction models. SPSS, MATLAB and other programming software are used to make the congestion index expression, and the internal relations among the influencing variables are discussed by combining the correlation coefficient matrix. Based on this, the probability distribution of congestion time is explored, and the results obtained have certain reference significance for formulating traffic policies, improving traffic trips and achieving the urbanization development goal based on smart growth.