Towards Standardising Reinforcement Learning Approaches for Production Scheduling Problems

Rinciog, Alexandru; Meyer, Anne Marie

doi:10.1016/j.procir.2022.05.117

Cited by 11 publications

(2 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, a wide range of so-called heuristic or metaheuristic methods [78] are developed, which trade off the solution quality with computational time. Although the heuristics and metaheuristics have improved to progressively mature since their emergence, the recent breakthrough of ML techniques and the increasing complexity of scheduling problems have aroused more interest in incorporating ML with traditional heuristic and metaheuristic methods for better performance [79,80].…”

Section: Job Shop Scheduling Problemsmentioning

confidence: 99%

A review on learning to solve combinatorial optimisation problems in manufacturing

Zhang

et al. 2023

IET Collab Intel Manufact

View full text Add to dashboard Cite

An efficient manufacturing system is key to maintaining a healthy economy today. With the rapid development of science and technology and the progress of human society, the modern manufacturing system is becoming increasingly complex, posing new challenges to both academia and industry. Ever since the beginning of industrialisation, leaps in manufacturing technology have always accompanied technological breakthroughs from other fields, for example, mechanics, physics, and computational science. Recently, machine learning (ML) technology, one of the crucial subjects of artificial intelligence, has made remarkable progress in many areas. This study thoroughly reviews how ML, specifically deep (reinforcement) learning, motivates new ideas for addressing challenging problems in manufacturing systems. We collect the literature targeting three aspects: scheduling, packing, and routing, which correspond to three pivotal cooperative production links of today's manufacturing system, that is, production, packing, and logistics respectively. For each aspect, we first present and discuss the state-of-the-art research. Then we summarise and analyse the development trends and point out future research opportunities and challenges. K E Y W O R D Sbin packing, combinatorial optimisation, deep reinforcement learning, job shop scheduling, manufacturing systems, vehicle routing | INTRODUCTIONCombinatorial optimisation problems (COPs), as one important branch of mathematical optimisation, have practical applications in many fields, such as communication, transportation, manufacturing and aroused broad research in industrial engineering, computer science, and operations research. Due to the NP (non-deterministic polynomial-time) hardness, finding their optimal solutions is challenging. In specific, the discrete solution space in COPs renders the optimisation less efficient, without the guidance of gradient as in continuous optimisation. Meanwhile, the complexity of searching the (near-)optimal solution(s) among feasible solutions could exponentially increase as the problem scale grows. Classic methods, including exact algorithms and (meta-)heuristics, generally depend on massive expertise and tuning work to solve specific problems. They are Cong Zhang, Yaoxin Wu, and Yining Ma are equal contribution.

show abstract

Section: Job Shop Scheduling Problemsmentioning

confidence: 99%

A review on learning to solve combinatorial optimisation problems in manufacturing

Zhang

et al. 2023

IET Collab Intel Manufact

View full text Add to dashboard Cite

show abstract

“…Therefore, a wide range of so-called construction heuristic or improvement heuristic methods [115] is developed, which trade off the solution quality with computational time. Although these heuristic algorithms have improved to progressively mature since their emergence, the recent breakthrough of machine learning (ML) techniques and the increasing complexity of scheduling problems have aroused more interest in incorporating ML with traditional heuristic and metaheuristic methods for better performance [116,117].…”

Section: Discussionmentioning

confidence: 99%

Intelligent job shop scheduling via deep reinforcement learning over graphs

Zhang¹

View full text Add to dashboard Cite

show abstract

Solving a Job Shop Scheduling Problem Using Q-Learning Algorithm

Belmamoune

Ghomri

Yahouni

2023

Studies in Computational Intelligence

View full text Add to dashboard Cite

Job Shop Scheduling Problem (JSSP) is among the combinatorial optimization and Non-Deterministic Polynomial-time (NP) problems. Researchers have contributed in this area using several methods, among the methods we have machine learning algorithms, more precisely Reinforcement Learning (RL). The reason why the scientists resort to RL is the adequacy of the algorithm for this type of problem. The results of the RL approach tend toward optimal or nearoptimal solutions. In this paper, we deal with the JSSP, using the RL algorithm, more specifically a Q-learning algorithm. We propose a new representation of the state of the environment. We introduce two evaluations of the agent using two different methods. The actions selected by the agent are the dispatching rules. Finally, we compared the results obtained by the approach with the literature.

show abstract

Towards Standardising Reinforcement Learning Approaches for Production Scheduling Problems

Cited by 11 publications

References 33 publications

A review on learning to solve combinatorial optimisation problems in manufacturing

A review on learning to solve combinatorial optimisation problems in manufacturing

Intelligent job shop scheduling via deep reinforcement learning over graphs

Solving a Job Shop Scheduling Problem Using Q-Learning Algorithm

Contact Info

Product

Resources

About