Causal inference-based graph neural network method for predicting asphalt pavement performance

To enhance the prediction accuracy of asphalt pavement rutting, this study introduces an end-to-end multivariate time series prediction model that integrates graph neural networks(GNN) with causal inference methodologies.The proposed model aims to effectively capture long-term and short-term tempora...

Full description

Saved in:
Bibliographic Details
Main Author: CHEN Kai;WANG Xiaohe;SHI Xinli;CAO Jinde
Format: Article
Language:English
Published: Editorial Department of Journal of Nantong University (Natural Science Edition) 2025-03-01
Series:Nantong Daxue xuebao. Ziran kexue ban
Subjects:
Online Access:https://ngzk.cbpt.cnki.net/portal/journal/portal/client/paper/aee504fdceeb10d65382bd95269b9489
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849713640906686464
author CHEN Kai;WANG Xiaohe;SHI Xinli;CAO Jinde
author_facet CHEN Kai;WANG Xiaohe;SHI Xinli;CAO Jinde
author_sort CHEN Kai;WANG Xiaohe;SHI Xinli;CAO Jinde
collection DOAJ
description To enhance the prediction accuracy of asphalt pavement rutting, this study introduces an end-to-end multivariate time series prediction model that integrates graph neural networks(GNN) with causal inference methodologies.The proposed model aims to effectively capture long-term and short-term temporal patterns as well as interdependencies among multiple variables. The model comprises four modules: global feature extraction, local feature extraction,causal inference, and dual-channel graph convolution. The global feature extraction module employs attention mechanisms and gated recurrent units(GRU) to capture long-term temporal dependencies within variables. The local feature extraction module utilizes dilated convolutional neural networks(CNN) with various kernel sizes to extract short-term temporal patterns at different scales. In the causal inference module, relationships among variables are identified using transfer entropy based on information theory, resulting in a relationship coefficient matrix that quantifies complex dependencies among variables. The dual-channel graph convolution module extends traditional low-pass graph convolutional neural networks by integrating a high-pass filter, simultaneously capturing low-frequency and high-frequency components of node signals or features to potentially improve prediction accuracy. The proposed approach was evaluated using the RIOHTrack dataset from the Research Institute of Highway Track, with comparisons conducted against several benchmark models, including the classical statistical model VARIMA, shallow learning model SVR, deep learning model GRU, attention mechanism-enhanced GRU, and TE-GCN. Experimental results indicate that the proposed model achieves superior predictive performance across various categories of asphalt pavement structures. Compared to traditional statistical models, deep learning-based models are more effective and stable, and the GRU module enhanced with attention mechanisms can capture long-term dependencies, further enhancing predictive performance. Overall, the proposed model provides a potentially effective solution for predicting asphalt pavement rutting and may offer practical insights for future pavement structure design and maintenance planning aimed at extending pavement lifespan.
format Article
id doaj-art-e6f44cfd97ba41ce8709e258ad0d57ff
institution DOAJ
issn 1673-2340
language English
publishDate 2025-03-01
publisher Editorial Department of Journal of Nantong University (Natural Science Edition)
record_format Article
series Nantong Daxue xuebao. Ziran kexue ban
spelling doaj-art-e6f44cfd97ba41ce8709e258ad0d57ff2025-08-20T03:13:54ZengEditorial Department of Journal of Nantong University (Natural Science Edition)Nantong Daxue xuebao. Ziran kexue ban1673-23402025-03-012401182710.12194/j.ntu.20240427001Causal inference-based graph neural network method for predicting asphalt pavement performanceCHEN Kai;WANG Xiaohe;SHI Xinli;CAO JindeTo enhance the prediction accuracy of asphalt pavement rutting, this study introduces an end-to-end multivariate time series prediction model that integrates graph neural networks(GNN) with causal inference methodologies.The proposed model aims to effectively capture long-term and short-term temporal patterns as well as interdependencies among multiple variables. The model comprises four modules: global feature extraction, local feature extraction,causal inference, and dual-channel graph convolution. The global feature extraction module employs attention mechanisms and gated recurrent units(GRU) to capture long-term temporal dependencies within variables. The local feature extraction module utilizes dilated convolutional neural networks(CNN) with various kernel sizes to extract short-term temporal patterns at different scales. In the causal inference module, relationships among variables are identified using transfer entropy based on information theory, resulting in a relationship coefficient matrix that quantifies complex dependencies among variables. The dual-channel graph convolution module extends traditional low-pass graph convolutional neural networks by integrating a high-pass filter, simultaneously capturing low-frequency and high-frequency components of node signals or features to potentially improve prediction accuracy. The proposed approach was evaluated using the RIOHTrack dataset from the Research Institute of Highway Track, with comparisons conducted against several benchmark models, including the classical statistical model VARIMA, shallow learning model SVR, deep learning model GRU, attention mechanism-enhanced GRU, and TE-GCN. Experimental results indicate that the proposed model achieves superior predictive performance across various categories of asphalt pavement structures. Compared to traditional statistical models, deep learning-based models are more effective and stable, and the GRU module enhanced with attention mechanisms can capture long-term dependencies, further enhancing predictive performance. Overall, the proposed model provides a potentially effective solution for predicting asphalt pavement rutting and may offer practical insights for future pavement structure design and maintenance planning aimed at extending pavement lifespan.https://ngzk.cbpt.cnki.net/portal/journal/portal/client/paper/aee504fdceeb10d65382bd95269b9489causal inferencemultivariate time series predictiondual-channel graph convolutional neural networkasphalt pavementrutting
spellingShingle CHEN Kai;WANG Xiaohe;SHI Xinli;CAO Jinde
Causal inference-based graph neural network method for predicting asphalt pavement performance
Nantong Daxue xuebao. Ziran kexue ban
causal inference
multivariate time series prediction
dual-channel graph convolutional neural network
asphalt pavement
rutting
title Causal inference-based graph neural network method for predicting asphalt pavement performance
title_full Causal inference-based graph neural network method for predicting asphalt pavement performance
title_fullStr Causal inference-based graph neural network method for predicting asphalt pavement performance
title_full_unstemmed Causal inference-based graph neural network method for predicting asphalt pavement performance
title_short Causal inference-based graph neural network method for predicting asphalt pavement performance
title_sort causal inference based graph neural network method for predicting asphalt pavement performance
topic causal inference
multivariate time series prediction
dual-channel graph convolutional neural network
asphalt pavement
rutting
url https://ngzk.cbpt.cnki.net/portal/journal/portal/client/paper/aee504fdceeb10d65382bd95269b9489
work_keys_str_mv AT chenkaiwangxiaoheshixinlicaojinde causalinferencebasedgraphneuralnetworkmethodforpredictingasphaltpavementperformance