The Application of ARIMA Model in Traffic Congestion Prediction in New York County

Abstract

Traffic congestion has become an important problem faced by large and medium-sized cities around the world. Previous studies have proposed a variety of models based on different research purposes and focuses, but there is still a lack of a unified understanding of the prediction of the comprehensive traffic congestion level of the main roads in New York City. To address this gap, this study collected traffic congestion data in the Manhattan area of New York City through the TomTom platform, analyzed it based on the ARIMA model, and compared the prediction results of the model with the actual congestion data to evaluate its applicability to the prediction of future traffic congestion levels. The results indicate that the model shows significant advantages in processing the relatively simple observation data in the region, and can provide more accurate prediction results of future traffic congestion, thus greatly reducing the difficulty of analysis and the time required for modeling, and finally making the model simple and accurate. The research results are expected to lay a foundation for future research on traffic congestion in New York City and provide a feasible and effective solution to the city's traffic problems.