Scholarlink

<p>This study explores the sustainable integration of ceramic waste into brick manufacturing, promoting efficient waste management while enhancing environmental sustainability. By diverting ceramic waste from landfills, the research demonstrates that bricks with ceramic waste exhibit superior mechanical properties. Specifically, bricks containing 40% ceramic waste powder (CWP) showed a 28% increase in compressive strength after 28 days, meeting Indian standards for porosity (10.33%) and water absorption. Beyond experimental validation, the study incorporates machine learning (ML) models—Support Vector Regression, Random Forest Regression, Gradient Boosting, XGBoost, and Multilayer Perceptron—to predict compressive strength. The Random Forest Regression model achieved the highest accuracy (R² = 0.96). A radar plot is used to compare model performance based on predictive accuracy, interpretability, and computational efficiency. This research combines experimental and computational approaches, paving the way for sustainable brick manufacturing by leveraging ceramic waste and advanced predictive modeling.</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 brick manufacturing: integrating ceramic waste and predictive analytics for sustainable production

  • 期刊论文

  • 发表日期：

    2025年12月4日

查看全文
  • Dilraj Preet Kaur,
  • Seema Raj,
  • Rupesh Kumar Tipu,
  • Siddarth Gupta,
  • Jyoti Sorout,
  • Pooja Lamba

摘要

This study explores the sustainable integration of ceramic waste into brick manufacturing, promoting efficient waste management while enhancing environmental sustainability. By diverting ceramic waste from landfills, the research demonstrates that bricks with ceramic waste exhibit superior mechanical properties. Specifically, bricks containing 40% ceramic waste powder (CWP) showed a 28% increase in compressive strength after 28 days, meeting Indian standards for porosity (10.33%) and water absorption. Beyond experimental validation, the study incorporates machine learning (ML) models—Support Vector Regression, Random Forest Regression, Gradient Boosting, XGBoost, and Multilayer Perceptron—to predict compressive strength. The Random Forest Regression model achieved the highest accuracy (R² = 0.96). A radar plot is used to compare model performance based on predictive accuracy, interpretability, and computational efficiency. This research combines experimental and computational approaches, paving the way for sustainable brick manufacturing by leveraging ceramic waste and advanced predictive modeling.