AI-Based Variable Importance Analysis of Mechanical and ASR Properties in Activated Waste Glass Mortar

Waste glass powder (WGP) faces challenges in recycling and regeneration, which is used as a partial substitute for concrete components, with its macro-mechanical properties being investigated. This study aims to elucidate the extent to which various variables affect the unconfined compressive streng...

Full description

Saved in:
Bibliographic Details
Main Authors: Fei Wu, Xin Zhang, Yanan Zhang, Dong Wang, Hua Tian, Jing Xu, Wei Luo, Yuzhuo Zhang
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Buildings
Subjects:
Online Access:https://www.mdpi.com/2075-5309/15/11/1866
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850158962828115968
author Fei Wu
Xin Zhang
Yanan Zhang
Dong Wang
Hua Tian
Jing Xu
Wei Luo
Yuzhuo Zhang
author_facet Fei Wu
Xin Zhang
Yanan Zhang
Dong Wang
Hua Tian
Jing Xu
Wei Luo
Yuzhuo Zhang
author_sort Fei Wu
collection DOAJ
description Waste glass powder (WGP) faces challenges in recycling and regeneration, which is used as a partial substitute for concrete components, with its macro-mechanical properties being investigated. This study aims to elucidate the extent to which various variables affect the unconfined compressive strength (UCS) and alkali–silica reactivity (ASR) of waste glass incorporated concrete. Initially, in the experimental procedure, 291 data points for the UCS and 485 data points for the ASR were obtained from laboratory tests. Subsequently, four machine learning models were introduced, including Gradient Boosting Regressor, Random Forest, Hist Gradient Boosting Regressor, and XGBoost. Their performance was analyzed and compared based on evaluation indexes. The findings reveal that Gradient Boosting Regressor accurately models the actual data distribution, generating reliable synthetic data. Partial dependence plots (PDPs) were used to understand the impact of individual features on glass concrete UCS and ASR, and Shapley additive explanation (SHAP) values were used to analyze the predictive output influenced by the contribution of each feature. The feature interaction effects analyzed through PDP indicate that UCS is highest when WGP is 202.5 kg/m<sup>3</sup>, and ASR is maximized when WGP is 708.75 kg/m<sup>3</sup>. The SHAP value analysis results reveal that the “alkali” feature exerts the most pronounced influence on the UCS model predictions. Conversely, in the case of the ASR model, the “curing duration” feature emerges as the primary driver of its predictions.
format Article
id doaj-art-e1a2e819711b4db2991f5f9efc662489
institution OA Journals
issn 2075-5309
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Buildings
spelling doaj-art-e1a2e819711b4db2991f5f9efc6624892025-08-20T02:23:44ZengMDPI AGBuildings2075-53092025-05-011511186610.3390/buildings15111866AI-Based Variable Importance Analysis of Mechanical and ASR Properties in Activated Waste Glass MortarFei Wu0Xin Zhang1Yanan Zhang2Dong Wang3Hua Tian4Jing Xu5Wei Luo6Yuzhuo Zhang7Jiangxi Traffic Engineering Assembly Manufacturing Co., Ltd., Nanchang 330000, ChinaChina Testing & Certification International Group Shanghai Co., Ltd., Shanghai 201203, ChinaSchool of Management, Shenyang Jianzhu University, Shenyang 110168, ChinaLiyang Market Comprehensive Inspection and Testing Center, Liyang 213300, ChinaChina Construction Fifth Engineering Bureau Co., Ltd., Changsha 410004, ChinaTongji University Library, Shanghai 200092, ChinaJiangxi Traffic Engineering Assembly Manufacturing Co., Ltd., Nanchang 330000, ChinaChina Testing & Certification International Group Shanghai Co., Ltd., Shanghai 201203, ChinaWaste glass powder (WGP) faces challenges in recycling and regeneration, which is used as a partial substitute for concrete components, with its macro-mechanical properties being investigated. This study aims to elucidate the extent to which various variables affect the unconfined compressive strength (UCS) and alkali–silica reactivity (ASR) of waste glass incorporated concrete. Initially, in the experimental procedure, 291 data points for the UCS and 485 data points for the ASR were obtained from laboratory tests. Subsequently, four machine learning models were introduced, including Gradient Boosting Regressor, Random Forest, Hist Gradient Boosting Regressor, and XGBoost. Their performance was analyzed and compared based on evaluation indexes. The findings reveal that Gradient Boosting Regressor accurately models the actual data distribution, generating reliable synthetic data. Partial dependence plots (PDPs) were used to understand the impact of individual features on glass concrete UCS and ASR, and Shapley additive explanation (SHAP) values were used to analyze the predictive output influenced by the contribution of each feature. The feature interaction effects analyzed through PDP indicate that UCS is highest when WGP is 202.5 kg/m<sup>3</sup>, and ASR is maximized when WGP is 708.75 kg/m<sup>3</sup>. The SHAP value analysis results reveal that the “alkali” feature exerts the most pronounced influence on the UCS model predictions. Conversely, in the case of the ASR model, the “curing duration” feature emerges as the primary driver of its predictions.https://www.mdpi.com/2075-5309/15/11/1866discarded glassmachine learningunconfined compressive strengthalkali–silica reactionpartial dependence plotsSHAP
spellingShingle Fei Wu
Xin Zhang
Yanan Zhang
Dong Wang
Hua Tian
Jing Xu
Wei Luo
Yuzhuo Zhang
AI-Based Variable Importance Analysis of Mechanical and ASR Properties in Activated Waste Glass Mortar
Buildings
discarded glass
machine learning
unconfined compressive strength
alkali–silica reaction
partial dependence plots
SHAP
title AI-Based Variable Importance Analysis of Mechanical and ASR Properties in Activated Waste Glass Mortar
title_full AI-Based Variable Importance Analysis of Mechanical and ASR Properties in Activated Waste Glass Mortar
title_fullStr AI-Based Variable Importance Analysis of Mechanical and ASR Properties in Activated Waste Glass Mortar
title_full_unstemmed AI-Based Variable Importance Analysis of Mechanical and ASR Properties in Activated Waste Glass Mortar
title_short AI-Based Variable Importance Analysis of Mechanical and ASR Properties in Activated Waste Glass Mortar
title_sort ai based variable importance analysis of mechanical and asr properties in activated waste glass mortar
topic discarded glass
machine learning
unconfined compressive strength
alkali–silica reaction
partial dependence plots
SHAP
url https://www.mdpi.com/2075-5309/15/11/1866
work_keys_str_mv AT feiwu aibasedvariableimportanceanalysisofmechanicalandasrpropertiesinactivatedwasteglassmortar
AT xinzhang aibasedvariableimportanceanalysisofmechanicalandasrpropertiesinactivatedwasteglassmortar
AT yananzhang aibasedvariableimportanceanalysisofmechanicalandasrpropertiesinactivatedwasteglassmortar
AT dongwang aibasedvariableimportanceanalysisofmechanicalandasrpropertiesinactivatedwasteglassmortar
AT huatian aibasedvariableimportanceanalysisofmechanicalandasrpropertiesinactivatedwasteglassmortar
AT jingxu aibasedvariableimportanceanalysisofmechanicalandasrpropertiesinactivatedwasteglassmortar
AT weiluo aibasedvariableimportanceanalysisofmechanicalandasrpropertiesinactivatedwasteglassmortar
AT yuzhuozhang aibasedvariableimportanceanalysisofmechanicalandasrpropertiesinactivatedwasteglassmortar