Mathematical modeling and hybrid evolutionary algorithm to schedule flexible job shop with discrete operation sequence flexibility DOI
Shuai Yuan, Xiaomin Zhu, Wei Cai

et al.

Computers & Operations Research, Journal Year: 2024, Volume and Issue: unknown, P. 106952 - 106952

Published: Dec. 1, 2024

Language: Английский

An improved squirrel search algorithm for a coefficient inversion problem of a singularly perturbed reaction-diffusion equation DOI Creative Commons
Xiongfa Mai,

Wenhe Bian,

Li‐Bin Liu

et al.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown

Published: Aug. 9, 2024

Abstract A novel numerical algorithm is proposed for solving a coefficient inverse problem of singularly perturbed reaction-diffusion equation with the final time observation data. Firstly, discretization scheme direct problem, barycentric form rational spectral approach based on sinh transformation used to discretize space derivative and Crank-Nicolson finite difference utilized approximate derivative. For construction we design new fitness function about reaction coefficient. Then integrating optimal neighborhood search strategy, random opposition learning strategy adaptive predator presence probability an improved squirrel named NOISSA proposed. Finally, 9 benchmark test functions are validate performance our NOISSA. Moreover, experiments carried out illustrate advantage in problems problems.

Language: Английский

Citations

0
Hybrid golden jackal and golden sine optimizer for tuning PID controllers DOI Creative Commons

Kailong Mou,

Ming Yang, Mengjian Zhang

et al.

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Sept. 27, 2024

Language: Английский

Citations

0
Dynamic scheduling for flexible job shop based on MachineRank algorithm and reinforcement learning DOI Creative Commons

Fujie Ren,

Haibin Liu

Scientific Reports, Journal Year: 2024, Volume and Issue: 14(1)

Published: Nov. 29, 2024

This paper investigates the Dynamic Flexible Job Shop Scheduling Problem (DFJSP), which is based on new job insertion, machine breakdowns, changes in processing time, and considering state of Automated Guided Vehicles (AGVs). The objective to minimize maximum completion time improve on-time rates. To address continuous production status learn most suitable actions (scheduling rules) at each rescheduling point, a Dueling Double Deep Q Network (D3QN) developed solve this problem. quality model solutions, MachineRank algorithm (MR) proposed, MR algorithm, seven composite scheduling rules are introduced. These aim select execute optimal operation an completed or disturbance occurs. Additionally, eight general features proposed represent point. By using as input D3QN, state-action values (Q-values) for rule can be obtained. Numerical experiments were conducted large number instances with different configurations, results demonstrated superiority generality D3QN compared various rules, other advanced standard Q-learning agents. effectiveness rationality dynamic trigger also validated.

Language: Английский

Citations

0
Mathematical modeling and hybrid evolutionary algorithm to schedule flexible job shop with discrete operation sequence flexibility DOI
Shuai Yuan, Xiaomin Zhu, Wei Cai

et al.

Computers & Operations Research, Journal Year: 2024, Volume and Issue: unknown, P. 106952 - 106952

Published: Dec. 1, 2024

Language: Английский

Citations

0