A multilogarithm linear regression traff ic congestion prediction model based on R language New York City's five asministrative districts (mainly in urban main roads and highways)

Abstract

Traffic congestion prediction plays a guiding role in vehicle route selection. Through traffic congestion prediction, it can not only help to reduce urban traffic pressure, but also facilitate vehicle travel and save vehicle travel costs.[1]Aiming at the problem of poor accuracy of estimated traffic congestion time caused by inaccurate estimation of vehicle speed in case of serious traffic congestion, people extracted the key factors of urban traffic congestion such as average travel speed, morning and evening peaks, average travel time, average delay time, average waiting time, and number of weeks. People constructed a multi-linear regression prediction model based on R language and used the selected ones and used the selected five administrative districts of New York City (mainly on the main roads of cities and highways), the data of road traffic conditions are monitored every five minutes for calculation and validity verification.