Traffic Prediction of Urban Intersections based on ARIMA Model

Abstract

In the urban planning, road traffic planning as an important part, through the formation of traffic layout network, can further accelerate the pace of urban development. However, the blockage will cause the car to start and stop regularly, and the car engine will emit a large amount of exhaust gas in these two stages, resulting in increased environmental pollution. Therefore, the focus needs to be on how to predict traffic flow efficiently. This paper focuses on the traffic flow density of intersections in one month, presents the change of traffic density by means of mapping, reasonably divides the training set and the test set, and uses ARIMA model to make effective prediction. It shows the changing relationship between traffic flow and time. Through reasonable prediction, the congestion problem is found in the first time, and at the same time, the solution measures are given, the traffic infrastructure is scientifically allocated, and the safety control system is set up, which provides reference for the later urban traffic development.