Scholarlink

<p>This study proposes a novel Surrogate-Assisted NSGA-III optimization framework to address the multi-objective scheduling problem in construction projects by simultaneously minimizing Project Completion Time (PCT), Project Completion Cost (PCC), and Project Rework Probability (PRP). Traditional time-cost optimization models often neglect the impact of rework, leading to incomplete decision-making in complex multi-mode environments. To overcome the computational inefficiencies of conventional evolutionary algorithms, the proposed approach integrates a Radial Basis Function (RBF) surrogate model with NSGA-III, significantly accelerating convergence without compromising solution quality. A 21-activity real-life case study with five execution modes per activity is presented, considering resource constraints and precedence relationships. The model generates 20 Pareto-optimal solutions, offering trade-offs among time, cost, and rework probability. Trade-off and sensitivity analyses, including correlation and clustering, reveal distinct scheduling strategies under varying project priorities. Comparative results demonstrate the superiority of the proposed model over existing methods such as MOPSO, OB-MODE, and LHS-based NSGA-III in terms of solution diversity, convergence, and computational efficiency. This framework equips project managers with a robust tool for data-driven, quality-aware construction scheduling under realistic constraints.</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

Surrogate-assisted NSGA-III optimization of time–cost–rework probability trade-offs in multi-mode construction project scheduling

  • 期刊论文

  • 发表日期：

    2025年11月12日

查看全文
  • Mani Bhushan,
  • Keshav Kumar,
  • Aviral Kumar,
  • Md Irfan Ansari,
  • Niraj Kumar Singh,
  • Subhash Chandra Bhushan,
  • Aditya Kumar

摘要

This study proposes a novel Surrogate-Assisted NSGA-III optimization framework to address the multi-objective scheduling problem in construction projects by simultaneously minimizing Project Completion Time (PCT), Project Completion Cost (PCC), and Project Rework Probability (PRP). Traditional time-cost optimization models often neglect the impact of rework, leading to incomplete decision-making in complex multi-mode environments. To overcome the computational inefficiencies of conventional evolutionary algorithms, the proposed approach integrates a Radial Basis Function (RBF) surrogate model with NSGA-III, significantly accelerating convergence without compromising solution quality. A 21-activity real-life case study with five execution modes per activity is presented, considering resource constraints and precedence relationships. The model generates 20 Pareto-optimal solutions, offering trade-offs among time, cost, and rework probability. Trade-off and sensitivity analyses, including correlation and clustering, reveal distinct scheduling strategies under varying project priorities. Comparative results demonstrate the superiority of the proposed model over existing methods such as MOPSO, OB-MODE, and LHS-based NSGA-III in terms of solution diversity, convergence, and computational efficiency. This framework equips project managers with a robust tool for data-driven, quality-aware construction scheduling under realistic constraints.