Scholarlink

<p>The construction industry faces escalating challenges in managing project schedules, mitigating risks, and achieving environmental sustainability, particularly in developing-country contexts where digital infrastructure, professional capacity, and building information modeling (BIM) regulatory frameworks remain limited. This study presents an integrated framework combining artificial intelligence (AI) with building information modeling (BIM) to optimize scheduling, risk assessment, and sustainability performance, validated across three real-world construction projects of varying scales located in Palestine: a high-rise residential tower, a mid-rise mixed-use building, and a smaller-scale mixed-use building. Structured quantitative datasets were extracted from BIM models developed using Autodesk Revit, Navisworks, and Primavera P6, and subsequently analyzed using machine learning (ML) algorithms, including Random Forest (RF) and XGBoost, alongside genetic algorithm (GA)-based material optimization. Baseline performance was established through conventional methods, namely the Critical Path Method (CPM), Program Evaluation and Review Technique (PERT), and Monte Carlo simulation, enabling systematic comparison against AI-based outputs. The results demonstrated measurable improvements across all three performance dimensions: Project durations were reduced by up to 3.52%, risk classification accuracy exceeded 99.7%, embodied carbon was reduced by 67%, and energy consumption decreased by 70% relative to conventional approaches. Feature importance analysis using Shapley Additive Explanations (SHAP) further enhanced the interpretability and transparency of AI-driven predictions, supporting practical adoption by construction project managers. The findings confirm that the proposed AI–BIM integration framework enables reliable, data-driven decision-making, demonstrates transferability across diverse geographical and resource-constrained contexts, and contributes to digital transformation and global carbon reduction objectives within the Construction 4.0 paradigm.</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

AI-Driven Optimization of Scheduling, Risk, and Sustainability in BIM-Enabled Construction Project Management

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月21日

查看全文
  • Suhib O. A. Amro,
  • Sepanta Naimi,
  • Abdullahi Abdu Ibrahim

摘要

The construction industry faces escalating challenges in managing project schedules, mitigating risks, and achieving environmental sustainability, particularly in developing-country contexts where digital infrastructure, professional capacity, and building information modeling (BIM) regulatory frameworks remain limited. This study presents an integrated framework combining artificial intelligence (AI) with building information modeling (BIM) to optimize scheduling, risk assessment, and sustainability performance, validated across three real-world construction projects of varying scales located in Palestine: a high-rise residential tower, a mid-rise mixed-use building, and a smaller-scale mixed-use building. Structured quantitative datasets were extracted from BIM models developed using Autodesk Revit, Navisworks, and Primavera P6, and subsequently analyzed using machine learning (ML) algorithms, including Random Forest (RF) and XGBoost, alongside genetic algorithm (GA)-based material optimization. Baseline performance was established through conventional methods, namely the Critical Path Method (CPM), Program Evaluation and Review Technique (PERT), and Monte Carlo simulation, enabling systematic comparison against AI-based outputs. The results demonstrated measurable improvements across all three performance dimensions: Project durations were reduced by up to 3.52%, risk classification accuracy exceeded 99.7%, embodied carbon was reduced by 67%, and energy consumption decreased by 70% relative to conventional approaches. Feature importance analysis using Shapley Additive Explanations (SHAP) further enhanced the interpretability and transparency of AI-driven predictions, supporting practical adoption by construction project managers. The findings confirm that the proposed AI–BIM integration framework enables reliable, data-driven decision-making, demonstrates transferability across diverse geographical and resource-constrained contexts, and contributes to digital transformation and global carbon reduction objectives within the Construction 4.0 paradigm.