Short-Term Traffic Speed Prediction For Freeways During Hurricane Evacuation: A Deep Learning Approach

Keywords

deep-learning; hurricane evacuation; Long short-term memory; time-series; traffic state

Abstract

Hurricane evacuation plays a critical role for effective disaster preparations. Giving accurate traffic prediction to evacuees enables a safe and smooth evacuation. Moreover, reliable traffic state prediction allows emergency managers to proactively respond to changes in traffic conditions. In this paper, we present a deep learning model to predict traffic speeds in freeways under extreme traffic demand, such as a hurricane evacuation. For prediction, we adopt a Long Short-Term Memory Neural Network (LSTM-NN) model. The approach is tested using real-world traffic data collected during hurricane Irma's evacuation for the interstate 75 (I-75), a major evacuation route in Florida. Using LSTM-NN, we perform several experiments for predicting speeds for 5 min, 10 min, and 15 min ahead of current time. The results are compared against other traditional prediction models such as KNN, ANN, ARIMA. We find that LSTM-NN performs better than these parametric and non-parametric models. The proposed method can be integrated with evacuation traffic management systems for a better evacuation operation.

Publication Date

12-7-2018

Publication Title

IEEE Conference on Intelligent Transportation Systems, Proceedings, ITSC

Volume

2018-November

Number of Pages

1291-1296

Document Type

Article; Proceedings Paper

Personal Identifier

scopus

DOI Link

https://doi.org/10.1109/ITSC.2018.8569443

Socpus ID

85060451290 (Scopus)

Source API URL

https://api.elsevier.com/content/abstract/scopus_id/85060451290

This document is currently not available here.

Share

COinS