A multi-layer CNN-GRUSKIP model based on transformer for spatial: temporal traffic flow prediction

Traffic flow prediction remains a cornerstone for intelligent transportation systems (ITS), influencing both route optimization and environmental efforts. While Recurrent Neural Networks (RNN) and traditional Convolutional Neural Networks (CNN) offer some insights into the spatial–temporal dynamics...

Full description

Bibliographic Details
Main Authors: Mohammad Ata, Karimeh Ibrahim, Hassan, Mohd Khair, Ismaeel, Ayad Ghany, Al-Haddad, Syed Abdul Rahman, Alquthami, Thamer‎, Alani, Sameer
Format: Article
Language:English
Published: Elsevier 2024
Online Access:http://psasir.upm.edu.my/id/eprint/114860/
http://psasir.upm.edu.my/id/eprint/114860/1/114860.pdf
_version_ 1848866617094569984
author Mohammad Ata, Karimeh Ibrahim
Hassan, Mohd Khair
Ismaeel, Ayad Ghany
Al-Haddad, Syed Abdul Rahman
Alquthami, Thamer‎
Alani, Sameer
author_facet Mohammad Ata, Karimeh Ibrahim
Hassan, Mohd Khair
Ismaeel, Ayad Ghany
Al-Haddad, Syed Abdul Rahman
Alquthami, Thamer‎
Alani, Sameer
author_sort Mohammad Ata, Karimeh Ibrahim
building UPM Institutional Repository
collection Online Access
description Traffic flow prediction remains a cornerstone for intelligent transportation systems (ITS), influencing both route optimization and environmental efforts. While Recurrent Neural Networks (RNN) and traditional Convolutional Neural Networks (CNN) offer some insights into the spatial–temporal dynamics of traffic data, they're often limited when navigating sparse and extended spatial–temporal patterns. In response, the CNN-GRUSKIP model emerges as a pioneering approach. Notably, it integrates the GRU-SKIP mechanism, a hybrid model that leverages the Gate Recurrent Unit's (GRU) capabilities to process sequences with the ’SKIP’ feature's ability to bypass and connect longer temporal dependencies, making it especially potent for traffic flow predictions with erratic and extended patterns. Another distinctive aspect is its non-standard 6-layer CNN, meticulously designed for in-depth spatiotemporal correlation extraction. The model comprises (1) the specialized CNN feature extraction, (2) the GRU-SKIP enhanced long-temporal module adept at capturing extended patterns, (3) a transformer module employing encoder-decoder and multi-attention mechanisms to hone prediction accuracy and trim model complexity, and (4) a bespoke prediction module. When tested against real-world datasets from California's Caltrans Performance Measurement System (PeMS), specifically PeMS districts 4 and 8, the CNN-GRUSKIP consistently outperformed established models such as ARIMA, Graph Wave Net, HA, LSTM, STGCN, and APTN. With its potent predictive prowess and adaptive architecture, the CNN-GRUSKIP model stands to redefine ITS applications, especially where nuanced traffic dynamics are in play. © 2024 THE AUTHORS
first_indexed 2025-11-15T14:23:27Z
format Article
id upm-114860
institution Universiti Putra Malaysia
institution_category Local University
language English
last_indexed 2025-11-15T14:23:27Z
publishDate 2024
publisher Elsevier
recordtype eprints
repository_type Digital Repository
spelling upm-1148602025-02-05T03:24:59Z http://psasir.upm.edu.my/id/eprint/114860/ A multi-layer CNN-GRUSKIP model based on transformer for spatial: temporal traffic flow prediction Mohammad Ata, Karimeh Ibrahim Hassan, Mohd Khair Ismaeel, Ayad Ghany Al-Haddad, Syed Abdul Rahman Alquthami, Thamer‎ Alani, Sameer Traffic flow prediction remains a cornerstone for intelligent transportation systems (ITS), influencing both route optimization and environmental efforts. While Recurrent Neural Networks (RNN) and traditional Convolutional Neural Networks (CNN) offer some insights into the spatial–temporal dynamics of traffic data, they're often limited when navigating sparse and extended spatial–temporal patterns. In response, the CNN-GRUSKIP model emerges as a pioneering approach. Notably, it integrates the GRU-SKIP mechanism, a hybrid model that leverages the Gate Recurrent Unit's (GRU) capabilities to process sequences with the ’SKIP’ feature's ability to bypass and connect longer temporal dependencies, making it especially potent for traffic flow predictions with erratic and extended patterns. Another distinctive aspect is its non-standard 6-layer CNN, meticulously designed for in-depth spatiotemporal correlation extraction. The model comprises (1) the specialized CNN feature extraction, (2) the GRU-SKIP enhanced long-temporal module adept at capturing extended patterns, (3) a transformer module employing encoder-decoder and multi-attention mechanisms to hone prediction accuracy and trim model complexity, and (4) a bespoke prediction module. When tested against real-world datasets from California's Caltrans Performance Measurement System (PeMS), specifically PeMS districts 4 and 8, the CNN-GRUSKIP consistently outperformed established models such as ARIMA, Graph Wave Net, HA, LSTM, STGCN, and APTN. With its potent predictive prowess and adaptive architecture, the CNN-GRUSKIP model stands to redefine ITS applications, especially where nuanced traffic dynamics are in play. © 2024 THE AUTHORS Elsevier 2024-12 Article PeerReviewed text en cc_by_nc_nd_4 http://psasir.upm.edu.my/id/eprint/114860/1/114860.pdf Mohammad Ata, Karimeh Ibrahim and Hassan, Mohd Khair and Ismaeel, Ayad Ghany and Al-Haddad, Syed Abdul Rahman and Alquthami, Thamer‎ and Alani, Sameer (2024) A multi-layer CNN-GRUSKIP model based on transformer for spatial: temporal traffic flow prediction. Ain Shams Engineering Journal, 15 (12). art. no. 103045. pp. 1-17. ISSN 2090-4479 https://www.sciencedirect.com/science/article/pii/S2090447924004209?via%3Dihub 10.1016/j.asej.2024.103045
spellingShingle Mohammad Ata, Karimeh Ibrahim
Hassan, Mohd Khair
Ismaeel, Ayad Ghany
Al-Haddad, Syed Abdul Rahman
Alquthami, Thamer‎
Alani, Sameer
A multi-layer CNN-GRUSKIP model based on transformer for spatial: temporal traffic flow prediction
title A multi-layer CNN-GRUSKIP model based on transformer for spatial: temporal traffic flow prediction
title_full A multi-layer CNN-GRUSKIP model based on transformer for spatial: temporal traffic flow prediction
title_fullStr A multi-layer CNN-GRUSKIP model based on transformer for spatial: temporal traffic flow prediction
title_full_unstemmed A multi-layer CNN-GRUSKIP model based on transformer for spatial: temporal traffic flow prediction
title_short A multi-layer CNN-GRUSKIP model based on transformer for spatial: temporal traffic flow prediction
title_sort multi-layer cnn-gruskip model based on transformer for spatial: temporal traffic flow prediction
url http://psasir.upm.edu.my/id/eprint/114860/
http://psasir.upm.edu.my/id/eprint/114860/
http://psasir.upm.edu.my/id/eprint/114860/
http://psasir.upm.edu.my/id/eprint/114860/1/114860.pdf