流水车间调度
计算机科学
进化算法
作业车间调度
数学优化
调度(生产过程)
遗传算法
进化计算
公平份额计划
算法设计
算法
进化规划
模因算法
人工智能
流量(数学)
分布式算法
动态优先级调度
分布式计算
两级调度
数学
作者
Fuqing Zhao,Nijiaer Sun,Ling Wang,Hongyan Sang
标识
DOI:10.1109/tevc.2026.3737767
摘要
The energy-aware distributed hybrid flow shop scheduling problem requires coordinated factory assignment, job sequencing, and machine allocation while minimizing total energy consumption (TEC) and makespan. Each true evaluation requires complete schedule decoding, making a fixed number of true schedule evaluations a limited search resource. This paper proposes a Budget-Aware Feasibility-Guided Surrogate-Assisted Evolutionary Algorithm (BF-SAEA) that coordinates surrogate-guided evaluation allocation and subsequent refinement through one global true-evaluation counter. A scale-specific multi-output surrogate and anchor-based trust region guide evaluation allocation, while a dynamically tightened excess-waiting guard provides auxiliary search guidance without changing the biobjective TEC–Cmaxformulation. Budget-controlled local and terminal refinement share the same evaluation resource, and only truly evaluated values are used for population and archive updating and for final performance assessment. Under a common budget of 3000 true evaluations, BF-SAEA significantly outperforms the classical evolutionary baselines and shows competitive performance against the recent scheduling-oriented algorithms.
科研通智能强力驱动
Strongly Powered by AbleSci AI