A Hybrid Evolutionary Algorithm Using Two Solution Representations for Hybrid Flow-Shop Scheduling Problem DOI
Jiaxin Fan, Yingli Li, Jin Xie

et al.

IEEE Transactions on Cybernetics, Journal Year: 2021, Volume and Issue: 53(3), P. 1752 - 1764

Published: Oct. 28, 2021

As an extension of the classical flow-shop scheduling problem, hybrid problem (HFSP) widely exists in large-scale industrial production systems and has been considered to be challenging for its complexity flexibility. Evolutionary algorithms based on encoding heuristic decoding approaches are shown effective solving HFSP. However, frequently used strategies can only search a limited area solution space, thus leading unsatisfactory performance during later period. In this article, evolutionary algorithm (HEA) using two representations is proposed solve HFSP makespan minimization. First, HEA searches space by permutation-based representation methods find some promising areas. Afterward, Tabu (TS) procedure disjunctive graph introduced expand searching further optimization. Two neighborhood structures focusing critical paths extended problem-specific backward schedules generate candidate solutions TS. The tested three public benchmark sets from existing literature, including 567 instances total, compared with state-of-the-art algorithms. Extensive experimental results indicate that performs much better than other Moreover, method finds new best 285 hard instances.

Language: Английский

A Self-Learning Discrete Jaya Algorithm for Multiobjective Energy-Efficient Distributed No-Idle Flow-Shop Scheduling Problem in Heterogeneous Factory System DOI
Fuqing Zhao, Ru Ma, Ling Wang

et al.

IEEE Transactions on Cybernetics, Journal Year: 2021, Volume and Issue: 52(12), P. 12675 - 12686

Published: Aug. 20, 2021

In this study, a self-learning discrete Jaya algorithm (SD-Jaya) is proposed to address the energy-efficient distributed no-idle flow-shop scheduling problem (FSP) in heterogeneous factory system (HFS-EEDNIFSP) with criteria of minimizing total tardiness (TTD), energy consumption (TEC), and load balancing (FLB). First, mixed-integer programming model HFS-EEDNIFSP presented. An evaluation criterion FLB combining completion time introduced. Second, operators selection strategy, which success rate each operator summarized as knowledge, designed for guiding operators. Third, energy-saving strategy reducing TEC. The FSP transformed be an permutation search idle times. speed operations adjacent are times reduced. effectiveness SD-Jaya tested on 60 benchmark instances. On quality solution, experimental results reveal that efficacy outperforms other algorithms addressing HFS-EEDNIFSP.

Language: Английский

Citations

173
Distributed scheduling problems in intelligent manufacturing systems DOI Open Access
Yaping Fu, Yushuang Hou, Zifan Wang

et al.

Tsinghua Science & Technology, Journal Year: 2021, Volume and Issue: 26(5), P. 625 - 645

Published: April 21, 2021

Currently, manufacturing enterprises face increasingly fierce market competition due to the various demands of customers and rapid development economic globalization. Hence, they have extend their production mode into distributed environments establish multiple factories in geographical locations. Nowadays, systems been widely adopted industrial processes. In recent years, many studies done on modeling optimization scheduling problems. This work provides a literature review problems intelligent systems. By summarizing evaluating existing problems, we analyze achievements current research status this field discuss ongoing studies. Insights regarding prior works are discussed uncover future directions, particularly swarm intelligence evolutionary algorithms, which used for managing focuses journal papers discovered using Google Scholar. After reviewing papers, work, trends point out some directions

Language: Английский

Citations

142
A Cooperative Memetic Algorithm With Learning-Based Agent for Energy-Aware Distributed Hybrid Flow-Shop Scheduling DOI
Jingjing Wang, Ling Wang

IEEE Transactions on Evolutionary Computation, Journal Year: 2021, Volume and Issue: 26(3), P. 461 - 475

Published: Aug. 19, 2021

With increasing environmental awareness and energy requirement, sustainable manufacturing has attracted growing attention. Meanwhile, distributed systems have become emerging due to the development of globalization. This article addresses energy-aware hybrid flow-shop scheduling (EADHFSP) with minimization makespan consumption simultaneously. We present a mixed-integer linear programming model propose cooperative memetic algorithm (CMA) reinforcement learning (RL)-based policy agent. First, an encoding scheme reasonable decoding method are designed, considering tradeoff between two conflicting objectives. Second, problem-specific heuristics presented for initialization generate diverse solutions. Third, solutions refined appropriate improvement operator selected by RL-based effective solution selection based on decomposition strategy is utilized balance convergence diversity. Fourth, intensification search multiple operators incorporated further enhance exploitation capability. Moreover, energy-saving strategies designed improving nondominated The effect parameter setting investigated extensive numerical tests carried out. comparative results demonstrate that special designs CMA superior existing algorithms in solving EADHFSP.

Language: Английский

Citations

137
Distributed Flow Shop Scheduling with Sequence-Dependent Setup Times Using an Improved Iterated Greedy Algorithm DOI Creative Commons
Han Xue, Yuyan Han, Qingda Chen

et al.

Complex System Modeling and Simulation, Journal Year: 2021, Volume and Issue: 1(3), P. 198 - 217

Published: Sept. 1, 2021

To meet the multi-cooperation production demand of enterprises, distributed permutation flow shop scheduling problem (DPFSP) has become frontier research in field manufacturing systems. In this paper, we investigate DPFSP by minimizing a makespan criterion under constraint sequence-dependent setup times. solve DPFSPs, significant developments some metaheuristic algorithms are necessary. context, simple and effective improved iterated greedy (NIG) algorithm is proposed to minimize DPFSPs. According features two-stage local search based on single job swapping block within key factory designed algorithm. We compare with state-of-the-art algorithms, including iterative (2019), Ruiz Pan discrete differential evolution (2018), artificial bee colony chemical reaction optimization (2017). Simulation results show that NIG outperforms compared algorithms.

Language: Английский

Citations

109
Novel MILP and CP models for distributed hybrid flowshop scheduling problem with sequence-dependent setup times DOI
Leilei Meng, Kaizhou Gao, Yaping Ren

et al.

Swarm and Evolutionary Computation, Journal Year: 2022, Volume and Issue: 71, P. 101058 - 101058

Published: March 25, 2022

Language: Английский

Citations

92
A novel shuffled frog-leaping algorithm with reinforcement learning for distributed assembly hybrid flow shop scheduling DOI
Jingcao Cai, Deming Lei, Jing Wang

et al.

International Journal of Production Research, Journal Year: 2022, Volume and Issue: 61(4), P. 1233 - 1251

Published: March 24, 2022

Distributed hybrid flow shop scheduling (DHFS) problem has attracted much attention in recent years; however, DHFS with actual processing constraints like assembly is seldom considered and reinforcement learning hardly embedded into meta-heuristic for DHFS. In this study, a distributed (DAHFS) fabrication, transportation mathematic model constructed. A new shuffled frog-learning algorithm Q-learning (QSFLA) proposed to minimise makespan. three-string representation used. newly defined process QSFLA select search strategy dynamically memeplex search. It composed of four actions based on the combination global search, neighbourhood solution acceptance rule, six states depicted by population evaluation elite diversity, reward function. number experiments are conducted. The computational results demonstrate that can provide promising DAHFS.

Language: Английский

Citations

79
An improved iterated greedy algorithm for the distributed hybrid flowshop scheduling problem DOI
Chao Lu, Jun Zheng, Lvjiang Yin

et al.

Engineering Optimization, Journal Year: 2023, Volume and Issue: 56(5), P. 792 - 810

Published: April 19, 2023

This study attempts to solve the distributed hybrid flowshop scheduling problem (DHFSP) with makespan criterion. First, a mixed-integer linear programming model for DHFSP is formulated. Then, an improved iterated greedy (IIG) algorithm developed handle this DHFSP. In IIG, new initialization strategy designed improve quality of initial solution. A operator, which combines perturbation operator and destruction/construction proposed enhance global search ability. According characteristics DHFSP, local method, integrates four neighbourhood structures, strengthen exploitation capability. The best parameter configuration IIG investigated through design experiments, validity each part verified by performing extensive experiments. Finally, compared other optimization algorithms on 100 large-scale instances. experimental results show that effective in addressing

Language: Английский

Citations

50
Energy-efficient distributed heterogeneous welding flow shop scheduling problem using a modified MOEA/D DOI
Guangchen Wang, Xinyu Li, Liang Gao

et al.

Swarm and Evolutionary Computation, Journal Year: 2021, Volume and Issue: 62, P. 100858 - 100858

Published: Feb. 10, 2021

Language: Английский

Citations

89
Multi-objective evolutionary algorithm based on multiple neighborhoods local search for multi-objective distributed hybrid flow shop scheduling problem DOI
Weishi Shao, Zhongshi Shao, Dechang Pi

et al.

Expert Systems with Applications, Journal Year: 2021, Volume and Issue: 183, P. 115453 - 115453

Published: June 19, 2021

Language: Английский

Citations

78
Improved Meta-Heuristics for Solving Distributed Lot-Streaming Permutation Flow Shop Scheduling Problems DOI
Yuxia Pan, Kaizhou Gao, Zhiwu Li

et al.

IEEE Transactions on Automation Science and Engineering, Journal Year: 2022, Volume and Issue: 20(1), P. 361 - 371

Published: Feb. 28, 2022

This paper addresses a distributed lot-streaming permutation flow shop scheduling problem that has various applications in real-life manufacturing systems. We aim to optimally assign jobs multiple factories and sequence them minimize the maximum completion time (Makespan). A mathematic model is first developed describe considered problem. Then, five meta-heuristics are executed solve it, including particle swarm optimization, genetic algorithm, harmony search, artificial bee colony, Jaya algorithm. To improve performance of these meta-heuristics, we employ Nawaz-Enscore-Ham (NEH) heuristic initialize populations propose improved strategies based on problem's feature. Finally, experiments carried out 120 instances. The verified. Comparisons discussions show colony algorithm with best competitiveness for solving proposed makespan criteria. Note Practitioners—In contemporary industry, traditional single-factory environment being replaced by multi-factory environment, as pattern can effectively production efficiency through reasonable resource allocation strategies. such significance practitioners. Although intelligent optimization provide an effective tool problems, most algorithms parameter-sensitive. challenge engineers parameter selection, which greatly impacts performance. ensure robustness algorithms, develop employing some Furthermore, setting test select appropriate values. As result, obtain schemes high-quality. It shown outperforms other well. methodology be readily applied real problems.

Language: Английский

Citations

68