Large-Scale Dynamic Scheduling for Flexible Job-Shop With Random Arrivals of New Jobs by Hierarchical Reinforcement Learning

Lei, Kun; Guo, Peng; Wang, Yi; Zhang, Jian; Xing, Meng; Qian, Linmao

doi:10.1109/tii.2023.3272661

Cited by 27 publications

(4 citation statements)

References 44 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A certain part of literature also addresses challenges associated with integrating new jobs into existing schedules [19], [20]; however, our work focuses on the scenario where machine breakdowns necessitate schedule adjustments without disrupting the established workflow on the shop floor. One notable literature review by [21] examines 140 related articles and presents an overview of mathematical models, integration frameworks, and qualitative analyses of different approaches in the context of Industry 4.0.…”

Section: B Real-time Dynamic Flexible Job Shop Scheduling Problem (Df...mentioning

confidence: 99%

An IoT-Enabled Real-Time Dynamic Scheduler for Flexible Job Shop Scheduling (FJSS) in an Industry 4.0-Based Manufacturing Execution System (MES 4.0)

Tariq,

Khan,

But

et al. 2024

IEEE Access

View full text Add to dashboard Cite

The most challenging aspect identified in this study revolves around effectively managing machine breakdowns to ensure uninterrupted production. This paper presents a real-time dynamic scheduling model that addresses the challenges of the Flexible Job Shop Scheduling Problem (FJSSP) while considering the occurrence of random machine breakdowns. An improved hybrid metaheuristic and rule-based multistrategy technique has been proposed that regenerates an optimized dynamic schedule when a random machine is interrupted. The proposed technique establishes that the presence of real-time system updates from IoT devices will improve scheduling decisions. The proposed methodology's efficacy is showcased through an extensive computational investigation encompassing 9 benchmark problems and a real-world case study, considering three performance objectives (Robustness, Stability and Compound Effectiveness). The results have been compared with three related techniques from literature. The proposed technique gives better results in most cases and can be adopted in increasing the performance of Manufacturing Execution Systems in an Industry 4.0 setup (MES 4.0).

show abstract

Section: B Real-time Dynamic Flexible Job Shop Scheduling Problem (Df...mentioning

confidence: 99%

An IoT-Enabled Real-Time Dynamic Scheduler for Flexible Job Shop Scheduling (FJSS) in an Industry 4.0-Based Manufacturing Execution System (MES 4.0)

Tariq,

Khan,

But

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Another type of RL algorithms is the actor-critic type and its variants, such as proximal policy optimization (PPO) [23]. The PPO algorithm has also been used in production scheduling problems [5,[24][25][26].…”

Section: Literature Reviewmentioning

confidence: 99%

“…A variant of the GNN is the graph isomorphism network (GIN), which is suitable for graph classification tasks [28]. Several studies have combined GIN and PPO to solve DFJSPs with a dynamic job arrival [5,24,25]. Zhang et al introduced the L2D framework [5], which combines the graph neural network (GNN) and proximal policy optimization (PPO) to solve JSSP.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Scheduling for the Flexible Job-Shop Problem with a Dynamic Number of Machines Using Deep Reinforcement Learning

Chang,

Liu,

You

2024

Information

View full text Add to dashboard Cite

The dynamic flexible job-shop problem (DFJSP) is a realistic and challenging problem that many production plants face. As the product line becomes more complex, the machines may suddenly break down or resume service, so we need a dynamic scheduling framework to cope with the changing number of machines over time. This issue has been rarely addressed in the literature. In this paper, we propose an improved learning-to-dispatch (L2D) model to generate a reasonable and good schedule to minimize the makespan. We formulate a DFJSP as a disjunctive graph and use graph neural networks (GINs) to embed the disjunctive graph into states for the agent to learn. The use of GINs enables the model to handle the dynamic number of machines and to effectively generalize to large-scale instances. The learning agent is a multi-layer feedforward network trained with a reinforcement learning algorithm, called proximal policy optimization. We trained the model on small-sized problems and tested it on various-sized problems. The experimental results show that our model outperforms the existing best priority dispatching rule algorithms, such as shortest processing time, most work remaining, flow due date per most work remaining, and most operations remaining. The results verify that the model has a good generalization capability and, thus, demonstrate its effectiveness.

show abstract

Design and calibration of a DRL algorithm for solving the job shop scheduling problem under unexpected job arrivals

Hammami,

Lardeux,

B. Hadj-Alouane

et al. 2024

Flex Serv Manuf J

View full text Add to dashboard Cite

Large-Scale Dynamic Scheduling for Flexible Job-Shop With Random Arrivals of New Jobs by Hierarchical Reinforcement Learning

Cited by 27 publications

References 44 publications

An IoT-Enabled Real-Time Dynamic Scheduler for Flexible Job Shop Scheduling (FJSS) in an Industry 4.0-Based Manufacturing Execution System (MES 4.0)

An IoT-Enabled Real-Time Dynamic Scheduler for Flexible Job Shop Scheduling (FJSS) in an Industry 4.0-Based Manufacturing Execution System (MES 4.0)

Scheduling for the Flexible Job-Shop Problem with a Dynamic Number of Machines Using Deep Reinforcement Learning

Design and calibration of a DRL algorithm for solving the job shop scheduling problem under unexpected job arrivals

Contact Info

Product

Resources

About