Scholarlink

<p>The rapid increase in demand for environmentally friendly construction materials has resulted in the exploration of alternative cementitious materials, such as Recycled Concrete Powder (RCP). This research presents a new framework for data-driven multi-objective optimization (MOO) to develop RCP-based Ultra High-Performance Concrete (UHPC) mixes that balance mechanical strength, economic feasibility, and environmental sustainability. Four predictive models were employed to optimize strength, sustainability, and cost, using predictive models trained on 380 mix designs that combined experimental and AI-generated data. XGBoost was shown to be the most precise model, achieving an R² score of 95.3%. The AI-optimized formulations maximized mechanical performance while simultaneously minimizing Fossil Fuel Depletion Potential (FFDP), Acidification Potential (ACDP), and Global Warming Potential (GWP). Compared with conventional UHPC, the optimized mixes reduced production cost by 60% and lowered environmental indicators by 55% to 65%. Multi-criteria decision-making (MCDM) further validated the robustness of the optimized solutions across different economic and ecological priorities. This framework provides a systematic pathway to lower the carbon footprint of UHPC through circular economy practices, offering scalable benefits for sustainable construction.</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

Integrated multi-criteria design and optimization of recycled concrete powder (RCP)-based ultra-high-performance concrete (UHPC)

  • 期刊论文

  • 发表日期：

    2026年2月14日

查看全文
  • AIB Farouk,
  • Salah U. Al-Dulaijan,
  • Mohammed A. Al-Huri,
  • Mohammed Fasil,
  • Yakubu Sani Wudil,
  • Mohammed A. Al-Osta

摘要

The rapid increase in demand for environmentally friendly construction materials has resulted in the exploration of alternative cementitious materials, such as Recycled Concrete Powder (RCP). This research presents a new framework for data-driven multi-objective optimization (MOO) to develop RCP-based Ultra High-Performance Concrete (UHPC) mixes that balance mechanical strength, economic feasibility, and environmental sustainability. Four predictive models were employed to optimize strength, sustainability, and cost, using predictive models trained on 380 mix designs that combined experimental and AI-generated data. XGBoost was shown to be the most precise model, achieving an R² score of 95.3%. The AI-optimized formulations maximized mechanical performance while simultaneously minimizing Fossil Fuel Depletion Potential (FFDP), Acidification Potential (ACDP), and Global Warming Potential (GWP). Compared with conventional UHPC, the optimized mixes reduced production cost by 60% and lowered environmental indicators by 55% to 65%. Multi-criteria decision-making (MCDM) further validated the robustness of the optimized solutions across different economic and ecological priorities. This framework provides a systematic pathway to lower the carbon footprint of UHPC through circular economy practices, offering scalable benefits for sustainable construction.