Assessment of AutoML frameworks for predicting compressive and flexural strength of recycled aggregate concrete

IF 7.9 3区 材料科学 Q1 GREEN & SUSTAINABLE SCIENCE & TECHNOLOGY
Deivid Campos , Bruno da Silva Macêdo , Zainab Al-Khafaji , Melike Aktaş Bozkurt , İhsan Erdem Kayral , Tiago Silveira Gontijo , Matteo Bodini , Camila M. Saporetti , Leonardo Goliatt
{"title":"Assessment of AutoML frameworks for predicting compressive and flexural strength of recycled aggregate concrete","authors":"Deivid Campos ,&nbsp;Bruno da Silva Macêdo ,&nbsp;Zainab Al-Khafaji ,&nbsp;Melike Aktaş Bozkurt ,&nbsp;İhsan Erdem Kayral ,&nbsp;Tiago Silveira Gontijo ,&nbsp;Matteo Bodini ,&nbsp;Camila M. Saporetti ,&nbsp;Leonardo Goliatt","doi":"10.1016/j.mtsust.2025.101200","DOIUrl":null,"url":null,"abstract":"<div><div>The use of recycled aggregate concrete (RAC) is crucial for promoting sustainable construction practices by mitigating the environmental impact associated with the extraction of natural aggregates (NA) and reducing <span><math><msub><mrow><mi>CO</mi></mrow><mrow><mn>2</mn></mrow></msub></math></span> emissions. This study aims to evaluate the performance of five automated machine learning (AutoML) frameworks — H2O, AutoKeras, FLAML, TPOT, and AutoGluon — in predicting the properties of RAC. The dataset comprises 638 samples with 13 variables, including compressive strength (CS) and flexural strength (FS). The results indicate that AutoKeras, based on deep learning, performed poorly due to the small dataset size and high dimensionality, which are not ideal for deep learning models. In contrast, FLAML and H2O demonstrated superior performance, with FLAML achieving the highest <span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span> (0.780) and lowest RMSE (6.928) for CS predictions. The Tukey test confirmed significant differences between AutoKeras and the other models, while AutoGluon, FLAML, H2O, and TPOT showed comparable effectiveness. This study highlights the importance of selecting appropriate AutoML models for accurate and reliable RAC property predictions, contributing to the reduction of <span><math><msub><mrow><mi>CO</mi></mrow><mrow><mn>2</mn></mrow></msub></math></span> emissions, conservation of natural resources, and promotion of a circular economy in the construction sector.</div></div>","PeriodicalId":18322,"journal":{"name":"Materials Today Sustainability","volume":"31 ","pages":"Article 101200"},"PeriodicalIF":7.9000,"publicationDate":"2025-08-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Materials Today Sustainability","FirstCategoryId":"88","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2589234725001290","RegionNum":3,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"GREEN & SUSTAINABLE SCIENCE & TECHNOLOGY","Score":null,"Total":0}
引用次数: 0

Abstract

The use of recycled aggregate concrete (RAC) is crucial for promoting sustainable construction practices by mitigating the environmental impact associated with the extraction of natural aggregates (NA) and reducing CO2 emissions. This study aims to evaluate the performance of five automated machine learning (AutoML) frameworks — H2O, AutoKeras, FLAML, TPOT, and AutoGluon — in predicting the properties of RAC. The dataset comprises 638 samples with 13 variables, including compressive strength (CS) and flexural strength (FS). The results indicate that AutoKeras, based on deep learning, performed poorly due to the small dataset size and high dimensionality, which are not ideal for deep learning models. In contrast, FLAML and H2O demonstrated superior performance, with FLAML achieving the highest R2 (0.780) and lowest RMSE (6.928) for CS predictions. The Tukey test confirmed significant differences between AutoKeras and the other models, while AutoGluon, FLAML, H2O, and TPOT showed comparable effectiveness. This study highlights the importance of selecting appropriate AutoML models for accurate and reliable RAC property predictions, contributing to the reduction of CO2 emissions, conservation of natural resources, and promotion of a circular economy in the construction sector.
预测再生骨料混凝土抗压和抗弯强度的AutoML框架评估
再生骨料混凝土(RAC)的使用对于促进可持续建筑实践至关重要,因为它减轻了与提取天然骨料(NA)相关的环境影响,并减少了二氧化碳的排放。本研究旨在评估五种自动机器学习(AutoML)框架(H2O、AutoKeras、FLAML、TPOT和AutoGluon)在预测RAC特性方面的性能。该数据集包括638个样本和13个变量,包括抗压强度(CS)和抗折强度(FS)。结果表明,基于深度学习的AutoKeras由于数据集规模小,维度高,不适合深度学习模型,因此表现不佳。相比之下,FLAML和H2O表现出更好的性能,FLAML在CS预测中获得最高的R2(0.780)和最低的RMSE(6.928)。Tukey测试证实AutoKeras与其他模型之间存在显著差异,而AutoGluon、FLAML、H2O和TPOT的效果相当。这项研究强调了选择合适的AutoML模型对准确可靠的RAC属性预测的重要性,有助于减少二氧化碳排放,保护自然资源,促进建筑部门的循环经济。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
CiteScore
5.80
自引率
6.40%
发文量
174
审稿时长
32 days
期刊介绍: Materials Today Sustainability is a multi-disciplinary journal covering all aspects of sustainability through materials science. With a rapidly increasing population with growing demands, materials science has emerged as a critical discipline toward protecting of the environment and ensuring the long term survival of future generations.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信