A Temporal Attention-Based SARIMA-BiLSTM Residual Learning Model Tuned by Grey Wolf Optimizer for Parallel Urban Traffic Forecasting

Accurate and timely traffic forecasting is essential for ensuring the efficiency, reliability, and safety of modern transportation networks. However, urban traffic exhibits high levels of temporal variability, spatial complexity, and nonlinear dependencies, posing challenges to traditional time seri...

Full description

Saved in:
Bibliographic Details
Main Authors: Manas Kamal Das, Christopher Columbus Chinnappan, E. Elakiya
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11083601/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850034630051233792
author Manas Kamal Das
Christopher Columbus Chinnappan
E. Elakiya
author_facet Manas Kamal Das
Christopher Columbus Chinnappan
E. Elakiya
author_sort Manas Kamal Das
collection DOAJ
description Accurate and timely traffic forecasting is essential for ensuring the efficiency, reliability, and safety of modern transportation networks. However, urban traffic exhibits high levels of temporal variability, spatial complexity, and nonlinear dependencies, posing challenges to traditional time series and machine learning models. In this study, a novel hybrid residual learning framework is proposed that integrates Seasonal Autoregressive Integrated Moving Average (SARIMA) with a Bidirectional Long Short-Term Memory (BiLSTM) network enhanced by a temporal attention mechanism. The SARIMA component captures underlying seasonal and linear patterns in traffic flow, while the residuals&#x2014;representing unmodeled nonlinear dynamics&#x2014;are learned by the BiLSTM module. To improve the model&#x2019;s generalization and predictive accuracy, a Grey Wolf Optimizer (GWO) is employed for automated hyperparameter tuning of both SARIMA and BiLSTM components. Furthermore, the entire pipeline is adapted for parallel execution to improve scalability and reduce computational latency, enabling faster processing of large-scale traffic data. Experimental evaluations on real-world traffic datasets demonstrate the effectiveness of the proposed model, achieving a processing speedup of <inline-formula> <tex-math notation="LaTeX">$10.11\times $ </tex-math></inline-formula> and a forecasting accuracy of 99.31%, outperforming conventional single-threaded and baseline hybrid models. The results indicate that the proposed attention-enhanced SARIMA-BiLSTM model, optimized via GWO and accelerated through parallel processing, provides a robust and scalable solution for real-time urban traffic forecasting applications.
format Article
id doaj-art-9af898144757442e8179ef38f1ada84f
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-9af898144757442e8179ef38f1ada84f2025-08-20T02:57:45ZengIEEEIEEE Access2169-35362025-01-011313607313608610.1109/ACCESS.2025.359010411083601A Temporal Attention-Based SARIMA-BiLSTM Residual Learning Model Tuned by Grey Wolf Optimizer for Parallel Urban Traffic ForecastingManas Kamal Das0Christopher Columbus Chinnappan1https://orcid.org/0000-0002-7525-1328E. Elakiya2https://orcid.org/0000-0003-3443-2840School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, IndiaSchool of Computer Science and Engineering, Vellore Institute of Technology, Chennai, IndiaSchool of Computer Science and Engineering, Vellore Institute of Technology, Chennai, IndiaAccurate and timely traffic forecasting is essential for ensuring the efficiency, reliability, and safety of modern transportation networks. However, urban traffic exhibits high levels of temporal variability, spatial complexity, and nonlinear dependencies, posing challenges to traditional time series and machine learning models. In this study, a novel hybrid residual learning framework is proposed that integrates Seasonal Autoregressive Integrated Moving Average (SARIMA) with a Bidirectional Long Short-Term Memory (BiLSTM) network enhanced by a temporal attention mechanism. The SARIMA component captures underlying seasonal and linear patterns in traffic flow, while the residuals&#x2014;representing unmodeled nonlinear dynamics&#x2014;are learned by the BiLSTM module. To improve the model&#x2019;s generalization and predictive accuracy, a Grey Wolf Optimizer (GWO) is employed for automated hyperparameter tuning of both SARIMA and BiLSTM components. Furthermore, the entire pipeline is adapted for parallel execution to improve scalability and reduce computational latency, enabling faster processing of large-scale traffic data. Experimental evaluations on real-world traffic datasets demonstrate the effectiveness of the proposed model, achieving a processing speedup of <inline-formula> <tex-math notation="LaTeX">$10.11\times $ </tex-math></inline-formula> and a forecasting accuracy of 99.31%, outperforming conventional single-threaded and baseline hybrid models. The results indicate that the proposed attention-enhanced SARIMA-BiLSTM model, optimized via GWO and accelerated through parallel processing, provides a robust and scalable solution for real-time urban traffic forecasting applications.https://ieeexplore.ieee.org/document/11083601/BiLSTMdeep learningGrey Wolf Optimizerhyperparameter tuningparallel computingresidual learning
spellingShingle Manas Kamal Das
Christopher Columbus Chinnappan
E. Elakiya
A Temporal Attention-Based SARIMA-BiLSTM Residual Learning Model Tuned by Grey Wolf Optimizer for Parallel Urban Traffic Forecasting
IEEE Access
BiLSTM
deep learning
Grey Wolf Optimizer
hyperparameter tuning
parallel computing
residual learning
title A Temporal Attention-Based SARIMA-BiLSTM Residual Learning Model Tuned by Grey Wolf Optimizer for Parallel Urban Traffic Forecasting
title_full A Temporal Attention-Based SARIMA-BiLSTM Residual Learning Model Tuned by Grey Wolf Optimizer for Parallel Urban Traffic Forecasting
title_fullStr A Temporal Attention-Based SARIMA-BiLSTM Residual Learning Model Tuned by Grey Wolf Optimizer for Parallel Urban Traffic Forecasting
title_full_unstemmed A Temporal Attention-Based SARIMA-BiLSTM Residual Learning Model Tuned by Grey Wolf Optimizer for Parallel Urban Traffic Forecasting
title_short A Temporal Attention-Based SARIMA-BiLSTM Residual Learning Model Tuned by Grey Wolf Optimizer for Parallel Urban Traffic Forecasting
title_sort temporal attention based sarima bilstm residual learning model tuned by grey wolf optimizer for parallel urban traffic forecasting
topic BiLSTM
deep learning
Grey Wolf Optimizer
hyperparameter tuning
parallel computing
residual learning
url https://ieeexplore.ieee.org/document/11083601/
work_keys_str_mv AT manaskamaldas atemporalattentionbasedsarimabilstmresiduallearningmodeltunedbygreywolfoptimizerforparallelurbantrafficforecasting
AT christophercolumbuschinnappan atemporalattentionbasedsarimabilstmresiduallearningmodeltunedbygreywolfoptimizerforparallelurbantrafficforecasting
AT eelakiya atemporalattentionbasedsarimabilstmresiduallearningmodeltunedbygreywolfoptimizerforparallelurbantrafficforecasting
AT manaskamaldas temporalattentionbasedsarimabilstmresiduallearningmodeltunedbygreywolfoptimizerforparallelurbantrafficforecasting
AT christophercolumbuschinnappan temporalattentionbasedsarimabilstmresiduallearningmodeltunedbygreywolfoptimizerforparallelurbantrafficforecasting
AT eelakiya temporalattentionbasedsarimabilstmresiduallearningmodeltunedbygreywolfoptimizerforparallelurbantrafficforecasting