Long term predictions for traffic forecasting

Long term predictions for traffic forecasting: How does the accuracy degrade with time?

Verlooy, Bas (TU Delft Electrical Engineering, Mathematics and Computer Science)

Congeduti, E. (mentor)
Iosifidis, G. (graduation committee)

Delft University of Technology

Computer Science and Engineering

CSE3000 Research Project

2023-06-28

Traffic prediction plays a big role in efficient transport planning capabilities and can reduce traffic congestion. In this study the application of Long Short-Term Memory (LSTM) models for predicting traffic volumes across varying prediction horizons is investigated. The data used is collected by the municipality of The Hague for a single month. The study focuses on comparing the performance of the LSTM across different time horizons up to 10 hours in the future. To evaluate the performance of the LSTM models, two common evaluation measures are employed: Root Mean Square Error (RMSE) and Symmetric Mean Absolute Percentage Error (SMAPE). The baseline for the predictions is set at a 15-minute future forecast. Comparing the 1-hour prediction against the 10-hour predictions relative to the baseline RMSE, the RMSE increased threefold. However, the SMAPE first increases, but surprisingly after 6 hours starts to decrease again.

http://resolver.tudelft.nl/uuid:e644a462-0276-4ab9-b012-7a1495bf3ebe

Student theses

bachelor thesis

© 2023 Bas Verlooy