Evolving Neural Networks to Solve a Two-Stage Hybrid Flow Shop Scheduling Problem with Family Setup Times

Lang, Sebastian; Reggelin, Tobias; Behrendt, Fabian; Nahhas, Abdulrahman

doi:10.24251/hicss.2020.160

Cited by 10 publications

(5 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It implies that A3C is able to adjust achieved policy to address new problems much better than PPO. To increase the confidence in the obtained results, we decided to conduct further analysis on published problem instances in [42] that are based on [21]. The authors investigated a two-stage HFS scheduling problem with family major and minor setup times.…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…The authors investigated a two-stage HFS scheduling problem with family major and minor setup times. The authors presented extensive analysis on the nature of the problem instance and presented an approach based on NeuroEvolution of Augmenting Topologies (NEAT) for solving the problems in [21]. Therefore, we built a simulation model for a two-stage production system and used the problem instance to further analyze the performance of the suggested approaches.…”

Section: Experiments and Resultsmentioning

confidence: 99%

“…However, we found limited research on the use of machine learning techniques for addressing scheduling problems of such complexity as we will discuss in the related works section. Nevertheless, some recent applications of machine learning techniques to address HFS scheduling problems are presented in [21]. The authors adopted NeuroEvolution of augmenting topologies to solve a two-stage HFS scheduling problem.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Deep Reinforcement Learning Techniques For Solving Hybrid Flow Shop Scheduling Problems: Proximal Policy Optimization (PPO) and Asynchronous Advantage Actor-Critic (A3C)

Nahhas¹,

Kharitonov²,

Turowski³

2022

Proceedings of the Annual Hawaii International Conference on System Sciences

Self Cite

View full text Add to dashboard Cite

Well-studied scheduling practices are fundamental for the successful support of core business processes in any manufacturing environment. Particularly, the Hybrid Flow Shop (HFS) scheduling problems are present in many manufacturing environments. The current advances in the field of Deep Reinforcement Learning (DRL) attracted the attention of both practitioners and academics to investigate their adoption beyond synthetic game-like applications. Therefore, we present an approach that is based on DRL techniques in conjunction with a discrete event simulation model to solve a real-world four-stage HFS scheduling problem. The main narrative behind the presented concepts is to expose a DRL agent to a gamelike environment using an indirect encoding. Two types of DRL techniques namely, Proximal Policy Optimization (PPO) and Asynchronous Advantage Actor-Critic (A3C), are evaluated for solving problems of different complexity. The computational results suggest that the DRL agents successfully learn appropriate policies for solving the investigated problem. In addition, the investigation shows that the agent can adjust their policies when we expose them to a different problem. We further evaluate the approach to solving problem instances published in the literature to establish a comparison.

show abstract

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Experiments and Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep Reinforcement Learning Techniques For Solving Hybrid Flow Shop Scheduling Problems: Proximal Policy Optimization (PPO) and Asynchronous Advantage Actor-Critic (A3C)

Nahhas¹,

Kharitonov²,

Turowski³

2022

Proceedings of the Annual Hawaii International Conference on System Sciences

Self Cite

View full text Add to dashboard Cite

show abstract

“…Since a matrix-structured system view also requires a certain degree of intelligence of the individual objects and stations, it is possible to control the production process via a predictive provision of materials (e.g. through the use of artificial intelligence methods, neural networks) [21]. The supply order for logistics and in particular the special requirements of Industrie 4.0 for the "8 Right of Logistics" (the right object at the right time in the right amount at the right place etc.)…”

Section: Resultsmentioning

confidence: 99%

Matrix Production Systems - Requirements and Influences on Logistics Planning for Decentralized Production Structures

Schmidtke¹,

Rettmann²,

Behrendt³

2021

Proceedings of the Annual Hawaii International Conference on System Sciences

Self Cite

View full text Add to dashboard Cite

In the context of the digital transformation of industry and within the framework of Industrie 4.0 and Factory Planning 4.0, new production-organizational principles with decentralized, modular and freely linked production cells are increasingly being discussed. The principle of matrix production with categorized and standardized workstations offers an extremely versatile production environment. It can be used to meet the challenge of an increasing number of product variants in variable quantities. This concept is predominantly considered from a theoretical point of view. Therefore, many aspects regarding the planning and operation of such systems are still up to debate. With the focus on logistics processes, this paper describes the requirements for such flexible, dynamic routing and self-organizing resources in material supply. Furthermore, they are investigated in a generic, conceptual model for a matrix production. Based on a reference scenario from the automotive industry, classical parameters from logistics and production organization are taken up. The influences with regard to decentralized material supply concepts and structural differences to flow production are shown by the results of simulation experiments with the generic model.

show abstract

“…As mentioned earlier, we will reconstruct the presented heuristic in [3] and compare its performance against the proposed framework. The most recent investigation of the problem is presented in [21]. The authors successfully applied neural networks for solving the problem and compared their results against the solutions presented in [3].…”

Section: Related Workmentioning

confidence: 99%

An adaptive scheduling framework for solving multi-objective hybrid flow shop scheduling problems

Nahhas

Krist²,

Turowski

2021

Proceedings of the Annual Hawaii International Conference on System Sciences

Self Cite

View full text Add to dashboard Cite

The proposed new technologies in the context of industry 4.0 challenge the current practices of scheduling in industry and their associated research in academia. The conventional optimization techniques that are employed for solving scheduling problems are either computationally expensive or lack the required quality. Therefore, in this paper, we propose an adaptive scheduling framework to address scheduling problems taking into account multi-objective optimality measures. The framework is motivated by a hybrid design to combine the use of heuristic and metaheuristic approaches. The main idea behind the presented concept is to achieve an acceptable tradeoff between the quality of the suggested solutions for a problem and the required computational effort to obtain them. The perused narrative in such implementation is combining some advantages of heuristic and metaheuristic approaches such as: the light execution time of heuristics and the robustness as well as the quality of metaheuristic approaches. The framework is evaluated for solving hybrid flow shop scheduling problems that are derived from a real use case.

show abstract

Evolving Neural Networks to Solve a Two-Stage Hybrid Flow Shop Scheduling Problem with Family Setup Times

Cited by 10 publications

References 19 publications

Deep Reinforcement Learning Techniques For Solving Hybrid Flow Shop Scheduling Problems: Proximal Policy Optimization (PPO) and Asynchronous Advantage Actor-Critic (A3C)

Deep Reinforcement Learning Techniques For Solving Hybrid Flow Shop Scheduling Problems: Proximal Policy Optimization (PPO) and Asynchronous Advantage Actor-Critic (A3C)

Matrix Production Systems - Requirements and Influences on Logistics Planning for Decentralized Production Structures

An adaptive scheduling framework for solving multi-objective hybrid flow shop scheduling problems

Contact Info

Product

Resources

About