An Estimation of Distribution Algorithm for Permutation Flow-Shop Scheduling Problem DOI Creative Commons
Sami Lemtenneche, Abdallah Bensayah, Abdelhakim Cheriet

et al.

Systems, Journal Year: 2023, Volume and Issue: 11(8), P. 389 - 389

Published: July 31, 2023

Estimation of distribution algorithms (EDAs) is a subset evolutionary widely used in various optimization problems, known for their favorable results. Each generation EDAs builds probabilistic model to represent the most promising individuals, and next created by sampling from this model. The primary challenge designing such lies effectively constructing mutual exclusivity constraint imposes an additional approach permutation-based problems. In study, we propose new EDA called Position-Guided Sampling Distribution Algorithm (PGS-EDA) specifically designed Unlike conventional approaches, our algorithm focuses on positions rather than elements during phase. We evaluate performance Permutation Flow-shop Scheduling Problem (PFSP). experiments conducted sizes Taillard instances provide evidence effectiveness addressing PFSP, particularly small medium-sized comparison results with other handle permutation problems demonstrate that PSG-EDA consistently achieves lowest Average Relative Percentage Deviation (ARPD) values 19 out 30 20 50 study. These findings validate superior terms minimizing makespan criterion PFSP.

Language: Английский

Hybrid Genetic and Spotted Hyena Optimizer for Flow Shop Scheduling Problem DOI Creative Commons
Toufik Mzili,

Ilyass Mzili,

Mohammed Essaid Riffi

et al.

Algorithms, Journal Year: 2023, Volume and Issue: 16(6), P. 265 - 265

Published: May 25, 2023

This paper presents a new hybrid algorithm that combines genetic algorithms (GAs) and the optimizing spotted hyena (SHOA) to solve production shop scheduling problem. The proposed GA-SHOA incorporates operators, such as uniform crossover mutation, into SHOA improve its performance. We evaluated on set of OR library instances compared it other state-of-the-art optimization algorithms, including SSO, SCE-OBL, CLS-BFO ACGA. experimental results show consistently finds optimal or near-optimal solutions for all tested instances, outperforming algorithms. Our contributes field in several ways. First, we propose effectively exploration exploitation capabilities SHO GA, resulting balanced efficient search process finding FSSP. Second, tailor GA methods specific requirements FSSP, encoding schemes, objective function evaluation constraint handling, which ensures is well suited address challenges posed by Third, perform comprehensive performance algorithm, demonstrating effectiveness terms solution quality computational efficiency. Finally, provide an in-depth analysis behavior discussing roles components their interactions during process, can help understand factors contributing success insight potential improvements adaptations combinatorial problems.

Language: Английский

Citations

12
A Mixed Integer Programming model to optimize production planning in the luxury textile industry DOI Open Access
Andrea Rossi, Lorenzo Tiacci,

Matteo Simonetti

et al.

Procedia Computer Science, Journal Year: 2025, Volume and Issue: 253, P. 1175 - 1184

Published: Jan. 1, 2025

Language: Английский

Citations

0
Integrated Optimisation of Shop Scheduling and Machine Layout for Discrete Manufacturing Considering Uncertain Events Based on an Improved Immune Genetic Algorithm DOI Creative Commons
Zhaoxi Hong, Yixiong Feng, Amir M. Fathollahi‐Fard

et al.

IET Collaborative Intelligent Manufacturing, Journal Year: 2025, Volume and Issue: 7(1)

Published: Jan. 1, 2025

ABSTRACT Shop scheduling and machine layout are two important aspects of discrete manufacturing. There strong coupling relationships between them, but they were conducted separately in the past, which significantly limits production performance improvement At same time, actual process workshop production, uncertain events not only often occur also may make existing schemes no longer suitable. To address such issues, integrated optimisation shop for manufacturing considering is proposed this paper, where minimum material handling cost, maximum space utilisation rate completion time selected as objectives. An improved immune genetic algorithm designed to solve corresponding mathematical model efficiently by dual‐layer encoding, good at global optimisation. Moreover, multistrategy redundancy‐aware rescheduling performed respond that regarded disturbances. The rationality superiority method verified a numerical case study wood–plastic composite materials with its layout, well under failures.

Language: Английский

Citations

0
Machine learning-driven task scheduling with dynamic K-means based clustering algorithm using fuzzy logic in FOG environment DOI Creative Commons

Muhammad Saad Sheikh,

Rabia Noor Enam,

Rehan Qureshi

et al.

Frontiers in Computer Science, Journal Year: 2023, Volume and Issue: 5

Published: Dec. 14, 2023

Fog Computing has emerged as a pivotal technology for enabling low-latency, context-aware, and efficient computing at the edge of network. Effective task scheduling plays vital role in optimizing performance fog systems. Traditional algorithms, primarily designed centralized cloud environments, often fail to cater dynamic, heterogeneous, resource-constrained nature nodes. To overcome these limitations, we introduce sophisticated machine learning-driven methodology that adapts allocation ever-changing environment's conditions. Our approach amalgamates K-Means clustering algorithm enhanced with fuzzy logic, robust unsupervised learning technique, efficiently group nodes based on their resource characteristics workload patterns. The proposed method combines capabilities K-means adaptability logic dynamically allocate tasks By leveraging techniques, demonstrate how can be intelligently allocated nodes, resulting reducing execution time, response time network usage. Through extensive experiments, showcase effectiveness our dynamic environments. Clustering proves time-effective identifying groups jobs per virtual (VM) efficiently. model evaluate approach, have utilized iFogSim. simulation results affirm showcasing significant enhancements reduction, minimized utilization, improved when compared existing non-machine methods within iFogSim framework.

Language: Английский

Citations

8
Research on Scheduling Algorithm of Knitting Production Workshop Based on Deep Reinforcement Learning DOI Creative Commons
Lei Sun, Weimin Shi,

Chang Xuan

et al.

Machines, Journal Year: 2024, Volume and Issue: 12(8), P. 579 - 579

Published: Aug. 22, 2024

Intelligent scheduling of knitting workshops is the key to realizing intelligent manufacturing. In view uncertainty workshop environment, it difficult for existing algorithms flexibly adjust strategies. This paper proposes a algorithm architecture based on deep reinforcement learning (DRL). First, problem represented by disjunctive graph, and mathematical model established. Then, multi-proximal strategy (multi-PPO) optimization training designed obtain optimal strategy, job selection machine are trained at same time. Finally, experimental platform built, proposed in this compared with common heuristic rules metaheuristic testing. The results show that superior solving problem, can achieve accuracy algorithm. addition, response speed excellent, which meets production needs has good guiding significance promoting

Language: Английский

Citations

1
Bio-Inspired Algorithms in Robotics Systems: An Overview DOI

Soukayna Belghiti Alaoui,

Badr El Kari,

Yassine Chaibi

et al.

Lecture notes in networks and systems, Journal Year: 2024, Volume and Issue: unknown, P. 496 - 513

Published: Jan. 1, 2024

Language: Английский

Citations

1
Multi-Objective Multi-Skill Resource-Constrained Project Scheduling Considering Flexible Resource Profiles DOI Creative Commons

Xu Luo,

Shunsheng Guo, Baigang Du

et al.

Applied Sciences, Journal Year: 2024, Volume and Issue: 14(5), P. 1921 - 1921

Published: Feb. 26, 2024

This paper addresses a novel multi-skill resource-constrained project scheduling problem with flexible resource profiles (F-MSRCPSP), in which the allocation of each activity consists certain number discrete resources and is allowed to be adjusted over its duration. The F-MSRCPSP aims, therefore, determine profile minimize make-span total cost simultaneously. Then, hybrid multi-objective fruit fly optimization algorithm proposed handle concerned problem. In algorithm, two parallel serial schedule generation schemes are introduced, aiming activities adjust allocated combinations. Additionally, heuristic strategies effectively select suitable combinations for activities. Moreover, series operators has been developed, including rejoining operator, empirical re-arrangement re-selection operator. These aim accelerate convergence speed enhance exploration algorithm. Finally, orthogonal test used optimal parameter combination, comparative experiments based on tests different scales conducted, along t-test. experimental results demonstrate that MOFOA-HS effective solving F-MSRCPSP.

Language: Английский

Citations

0
An Estimation of Distribution Algorithm for Permutation Flow-Shop Scheduling Problem DOI Creative Commons
Sami Lemtenneche, Abdallah Bensayah, Abdelhakim Cheriet

et al.

Systems, Journal Year: 2023, Volume and Issue: 11(8), P. 389 - 389

Published: July 31, 2023

Estimation of distribution algorithms (EDAs) is a subset evolutionary widely used in various optimization problems, known for their favorable results. Each generation EDAs builds probabilistic model to represent the most promising individuals, and next created by sampling from this model. The primary challenge designing such lies effectively constructing mutual exclusivity constraint imposes an additional approach permutation-based problems. In study, we propose new EDA called Position-Guided Sampling Distribution Algorithm (PGS-EDA) specifically designed Unlike conventional approaches, our algorithm focuses on positions rather than elements during phase. We evaluate performance Permutation Flow-shop Scheduling Problem (PFSP). experiments conducted sizes Taillard instances provide evidence effectiveness addressing PFSP, particularly small medium-sized comparison results with other handle permutation problems demonstrate that PSG-EDA consistently achieves lowest Average Relative Percentage Deviation (ARPD) values 19 out 30 20 50 study. These findings validate superior terms minimizing makespan criterion PFSP.

Language: Английский

Citations

0