Scholarlink

<p>This study explores optimizing concrete mix designs by incorporating agricultural waste by-products to promote sustainability in the construction industry. Data were collected on parameters such as by-product content, compressive strength, tensile strength, sulfate resistance, chloride ion penetration, water-cement ratio, curing duration, and slump. Machine learning models, including Linear Regression, Random Forest, Gradient Boosting, Support Vector Regression, and Artificial Neural Networks, were applied to predict these properties. The models were evaluated using metrics such as Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), and the coefficient of determination (R²). Sequential Least Squares Quadratic Programming (SLSQP) was used to optimize the mix designs. The study aimed to identify the optimal by-product type and percentage that maximizes strength and performance. Results indicated that Corn Cob Ash (15–20%) is optimal for early strength development, while Palm Oil Fuel Ash, Rice Husk Ash, Sugarcane Bagasse Ash, and Wheat Straw Ash are suitable for mid- and later-stage strength. This research presents a machine learning-based approach to optimizing concrete mixes, reducing the need for extensive experimental testing while advancing sustainable construction practices.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Optimizing Agricultural Waste By-Products: A Machine Learning Approach for Sustainable Construction Practices

  • 期刊论文

  • 发表日期：

    2025年2月6日

查看全文
  • Pradyut Anand,
  • Surya Dev Singh,
  • Priyam Nath Bhowmik,
  • Veeresh Boya,
  • Shatrudhan Pandey

摘要

This study explores optimizing concrete mix designs by incorporating agricultural waste by-products to promote sustainability in the construction industry. Data were collected on parameters such as by-product content, compressive strength, tensile strength, sulfate resistance, chloride ion penetration, water-cement ratio, curing duration, and slump. Machine learning models, including Linear Regression, Random Forest, Gradient Boosting, Support Vector Regression, and Artificial Neural Networks, were applied to predict these properties. The models were evaluated using metrics such as Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), and the coefficient of determination (R²). Sequential Least Squares Quadratic Programming (SLSQP) was used to optimize the mix designs. The study aimed to identify the optimal by-product type and percentage that maximizes strength and performance. Results indicated that Corn Cob Ash (15–20%) is optimal for early strength development, while Palm Oil Fuel Ash, Rice Husk Ash, Sugarcane Bagasse Ash, and Wheat Straw Ash are suitable for mid- and later-stage strength. This research presents a machine learning-based approach to optimizing concrete mixes, reducing the need for extensive experimental testing while advancing sustainable construction practices.