01764350

Component

https://annualmeeting.myt...nlineProgram/Details/15715

Traffic prediction is an essential component in intelligent transportation systems. Various methods have been developed to solve this challenging problem over the years, including time series models, regression models, and, more recently, deep learning models. This paper provides an unbiased comparison of these methods under a variety of settings and also addresses the critical question of whether deep learning approaches can offer significant improvements over classical machine learning methods. The authors used a traffic simulation model of the Greater Toronto Area to generate traffic data for a stretch of highway as well as an urban region. Using these datasets, the authors compared the methods under five scenarios with different prediction horizons, the presence of missing data, and the presence of traffic events unseen in the training data. The authors' experimental results show that deep learning methods of traffic prediction, including graph convolutional neural networks, are effective for traffic prediction. Graph convolutional neural networks with shared parameters are very compact, resistant to overfitting, and performed well in all of the authors' experiments. However, ensemble methods such as random forest regression can generate more accurate predictions at the cost of higher resource consumption during training, which may become a challenge in large transportation networks. Overall, the authors found that deep learning architectures should be carefully designed by restricting the input to features with known influences on the predictions, which can guide parameter learning and improve performance.

This paper was sponsored by TRB committee AED50 Standing Committee on Artificial Intelligence and Advanced Computing Applications.

TRBAM-21-02099

English

21p

2021

Transportation Research Board 100th Annual Meeting

Web

Figures; References; Tables

Alternatives analysis; Evaluation; Intelligent transportation systems; Machine learning; Mathematical prediction; Neural networks; Traffic data

Highways; Operations and Traffic Management

Transportation Research Board Annual Meeting 2021 Paper #TRBAM-21-02099

TRIS, TRB, ATRI

Dec 23 2020 11:27AM