Construction and Empirical Study of Short-term Prediction Model for Urban Road Speed

DOI

10.2991/978-94-6463-986-5_45How to use a DOI?

Keywords

Intelligent transportation; Deep learning; Random Forest

Abstract

This study addresses short-term traffic speed prediction for urban road networks, a critical component of intelligent transportation systems for easing congestion, optimizing traffic signal control, and improving travel efficiency. Using real-world traffic speed data, the performance of a Long Short-Term Memory deep learning model is compared with three traditional approaches: Naive Moving Average, Linear Regression, and Random Forest. All models are evaluated on the same test set using Mean Absolute Error, Root Mean Squared Error, and coefficient of determination. Experimental results show that the LSTM model achieves the highest accuracy, with MAE = 0.49 km/h, RMSE = 0.61 km/h, and R2 = 0.9321, significantly outperforming Naive Moving Average (MAE = 1.95, RMSE = 2.466, R2 = 0.1109), Linear Regression (MAE = 0.96, RMSE = 1.169, R2 = 0.8004), and Random Forest (MAE = 1.10, RMSE = 1.29, R2 = 0.7554). These findings demonstrate that LSTM not only accurately captures overall speed trends but also adapts well to short-term fluctuations, highlighting its potential for real-time traffic forecasting and congestion mitigation in urban environments.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Kaize Liu
PY  - 2026
DA  - 2026/02/18
TI  - Construction and Empirical Study of Short-term Prediction Model for Urban Road Speed
BT  - Proceedings of the 2025 International Conference on Electronics, Electrical and Grid Technology (ICEEGT 2025)
PB  - Atlantis Press
SP  - 441
EP  - 449
SN  - 2352-5401
UR  - https://doi.org/10.2991/978-94-6463-986-5_45
DO  - 10.2991/978-94-6463-986-5_45
ID  - Liu2026
ER  -