A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
XGBoost-SHAP framework for asphalt pavement condition evaluation
https://orcid.org/0000-0002-5087-7308
https://orcid.org/0000-0001-7997-1421
Abstract This study addresses the challenge of enhancing pavement condition assessment methodologies by proposing a universal XGBoost-SHAP framework. Leveraging diverse numerical input variables, including cracking, plasticity index, maximum dry density, California bearing ratio, soil type, and layer thickness, this framework aims to derive pivotal pavement condition parameters efficiently. The research demonstrates the framework’s efficacy in facilitating data-driven decision-making, offering a cost-effective alternative to traditional falling weight deflectometer (FWD) testing. Notably, the study utilizes a dataset of 2001 instances from the non-core road network of Andhra Pradesh State for model training and validation. Results reveal the clear advantages of the XGBoost-SHAP model over conventional FWD approach, particularly in terms of cost-efficiency, transparency, and precision. Detailed analysis employing Shapley Additive Explanations (SHAP) identifies cracking percentage as a key predictor for surface condition parameters, while California Bearing Ratio (CBR) emerges as crucial for deflection ratio prediction, highlighting the model’s predictive power and transparency. Among all the ensemble approaches including Random Forest, XGBoost, Light GBM, and other ML algorithms, XGBoost exhibits the highest , the lowest MSE and MAE, and an extremely low MAPE, demonstrating its superior prediction accuracy. Overall, this research introduces a promising avenue for advancing pavement condition assessment, offering an economically viable, data-centric solution characterized by heightened accuracy and transparency. By bridging the gap between traditional methodologies and advanced machine learning techniques, the proposed framework holds promise for revolutionizing pavement management practices.
Highlights This study introduces a cost-effective and transparent XGBoost-SHAP framework, outperforming traditional falling weight deflectometer (FWD) methods. It effectively computes crucial pavement condition parameters, including surface and base curvature indices, damage indices, deflection ratio, and pavement profile area. Among machine learning algorithms, XGBoost exhibits the highest accuracy in predicting five key pavement parameters, making it an ideal choice for pavement quality assessment. Shapley Additive Explanations (SHAP) analysis reveals that pavement cracking percentage significantly impacts surface and base condition indices, while California Bearing Ratio (CBR) dominates deflection ratio prediction. This research others an economical, data-centric solution for pavement condition evaluation, addressing the challenges of extensive road networks and costly FWD testing. The XGBoost-SHAP model enhances accuracy and transparency in pavement management.
Graphical abstract Display Omitted
XGBoost-SHAP framework for asphalt pavement condition evaluation
https://orcid.org/0000-0002-5087-7308
https://orcid.org/0000-0001-7997-1421
Abstract This study addresses the challenge of enhancing pavement condition assessment methodologies by proposing a universal XGBoost-SHAP framework. Leveraging diverse numerical input variables, including cracking, plasticity index, maximum dry density, California bearing ratio, soil type, and layer thickness, this framework aims to derive pivotal pavement condition parameters efficiently. The research demonstrates the framework’s efficacy in facilitating data-driven decision-making, offering a cost-effective alternative to traditional falling weight deflectometer (FWD) testing. Notably, the study utilizes a dataset of 2001 instances from the non-core road network of Andhra Pradesh State for model training and validation. Results reveal the clear advantages of the XGBoost-SHAP model over conventional FWD approach, particularly in terms of cost-efficiency, transparency, and precision. Detailed analysis employing Shapley Additive Explanations (SHAP) identifies cracking percentage as a key predictor for surface condition parameters, while California Bearing Ratio (CBR) emerges as crucial for deflection ratio prediction, highlighting the model’s predictive power and transparency. Among all the ensemble approaches including Random Forest, XGBoost, Light GBM, and other ML algorithms, XGBoost exhibits the highest , the lowest MSE and MAE, and an extremely low MAPE, demonstrating its superior prediction accuracy. Overall, this research introduces a promising avenue for advancing pavement condition assessment, offering an economically viable, data-centric solution characterized by heightened accuracy and transparency. By bridging the gap between traditional methodologies and advanced machine learning techniques, the proposed framework holds promise for revolutionizing pavement management practices.
Highlights This study introduces a cost-effective and transparent XGBoost-SHAP framework, outperforming traditional falling weight deflectometer (FWD) methods. It effectively computes crucial pavement condition parameters, including surface and base curvature indices, damage indices, deflection ratio, and pavement profile area. Among machine learning algorithms, XGBoost exhibits the highest accuracy in predicting five key pavement parameters, making it an ideal choice for pavement quality assessment. Shapley Additive Explanations (SHAP) analysis reveals that pavement cracking percentage significantly impacts surface and base condition indices, while California Bearing Ratio (CBR) dominates deflection ratio prediction. This research others an economical, data-centric solution for pavement condition evaluation, addressing the challenges of extensive road networks and costly FWD testing. The XGBoost-SHAP model enhances accuracy and transparency in pavement management.
Graphical abstract Display Omitted
XGBoost-SHAP framework for asphalt pavement condition evaluation
https://orcid.org/0000-0002-5087-7308
https://orcid.org/0000-0001-7997-1421
Abstract This study addresses the challenge of enhancing pavement condition assessment methodologies by proposing a universal XGBoost-SHAP framework. Leveraging diverse numerical input variables, including cracking, plasticity index, maximum dry density, California bearing ratio, soil type, and layer thickness, this framework aims to derive pivotal pavement condition parameters efficiently. The research demonstrates the framework’s efficacy in facilitating data-driven decision-making, offering a cost-effective alternative to traditional falling weight deflectometer (FWD) testing. Notably, the study utilizes a dataset of 2001 instances from the non-core road network of Andhra Pradesh State for model training and validation. Results reveal the clear advantages of the XGBoost-SHAP model over conventional FWD approach, particularly in terms of cost-efficiency, transparency, and precision. Detailed analysis employing Shapley Additive Explanations (SHAP) identifies cracking percentage as a key predictor for surface condition parameters, while California Bearing Ratio (CBR) emerges as crucial for deflection ratio prediction, highlighting the model’s predictive power and transparency. Among all the ensemble approaches including Random Forest, XGBoost, Light GBM, and other ML algorithms, XGBoost exhibits the highest , the lowest MSE and MAE, and an extremely low MAPE, demonstrating its superior prediction accuracy. Overall, this research introduces a promising avenue for advancing pavement condition assessment, offering an economically viable, data-centric solution characterized by heightened accuracy and transparency. By bridging the gap between traditional methodologies and advanced machine learning techniques, the proposed framework holds promise for revolutionizing pavement management practices.
Highlights This study introduces a cost-effective and transparent XGBoost-SHAP framework, outperforming traditional falling weight deflectometer (FWD) methods. It effectively computes crucial pavement condition parameters, including surface and base curvature indices, damage indices, deflection ratio, and pavement profile area. Among machine learning algorithms, XGBoost exhibits the highest accuracy in predicting five key pavement parameters, making it an ideal choice for pavement quality assessment. Shapley Additive Explanations (SHAP) analysis reveals that pavement cracking percentage significantly impacts surface and base condition indices, while California Bearing Ratio (CBR) dominates deflection ratio prediction. This research others an economical, data-centric solution for pavement condition evaluation, addressing the challenges of extensive road networks and costly FWD testing. The XGBoost-SHAP model enhances accuracy and transparency in pavement management.
Graphical abstract Display Omitted
XGBoost-SHAP framework for asphalt pavement condition evaluation
Gupta, Aakash (author) / Gowda, Sachin (author) / Tiwari, Achyut (author) / Gupta, Ashok Kumar (author)
2024-04-06
Article (Journal)
Electronic Resource
English
XGBoost-SHAP framework for asphalt pavement condition evaluation
| Elsevier | 2024
Subsurface Condition Evaluation of Asphalt Pavement for Pavement Preservation Treatments
| British Library Online Contents | 2016