Reinforcement Learning approaches for the Parallel Machines Job Shop Scheduling Problem

Yailen Martinez Jimenez; Wauters Tony; Ann Nowe; Katja Verbeeck; De Causmaecker Patrick; Juliett Suarez; Rafael Bello

Reinforcement Learning approaches for the Parallel Machines Job Shop Scheduling Problem

Yailen Martinez Jimenez, Wauters Tony, Ann Nowe, Katja Verbeeck, De Causmaecker Patrick, Juliett Suarez, Rafael Bello

Research output: Chapter in Book/Report/Conference proceeding › Conference paper

Abstract

This paper presents two Reinforcement Learning approaches for the Parallel Machines Job Shop Scheduling Problem. This NP-hard optimization problem where operations have to be assigned to machines of different workcenters, each workcenter having a number of identical parallel machines, is hard to solve and is therefore tackled with learning based methods. The objective used is the minimization of the schedule makespan. Two viewpoints are taken, one where resources are intelligent agents and have to choose what operation to process next, and another where operations themselves are seen as the agents that have to choose their mutual scheduling order. As Reinforcement Learning methods we use a value iteration method (Q-Learning) and a policy iteration method
(Learning Automata). The results of both approaches improve on recent published results from the literature.

Original language	English
Title of host publication	Proceedings of the Cuba-Flanders Workshop on Machine Learning and Knowledge Discovery
Publication status	Published - Feb 2010

Publication series

Name	Proceedings of the Cuba-Flanders Workshop on Machine Learning and Knowledge Discovery

Keywords

Reinforcement Learning
Job Shop Scheduling
Parallel Machines Job Shop Scheduling

Cite this

Martinez Jimenez, Y., Tony, W., Nowe, A., Verbeeck, K., Patrick, D. C., Suarez, J., & Bello, R. (2010). Reinforcement Learning approaches for the Parallel Machines Job Shop Scheduling Problem. In Proceedings of the Cuba-Flanders Workshop on Machine Learning and Knowledge Discovery (Proceedings of the Cuba-Flanders Workshop on Machine Learning and Knowledge Discovery).

Martinez Jimenez, Yailen ; Tony, Wauters ; Nowe, Ann ; Verbeeck, Katja ; Patrick, De Causmaecker ; Suarez, Juliett ; Bello, Rafael. / Reinforcement Learning approaches for the Parallel Machines Job Shop Scheduling Problem. Proceedings of the Cuba-Flanders Workshop on Machine Learning and Knowledge Discovery. 2010. (Proceedings of the Cuba-Flanders Workshop on Machine Learning and Knowledge Discovery).

@inproceedings{ed51162a603f4810b0e40a693113443c,

title = "Reinforcement Learning approaches for the Parallel Machines Job Shop Scheduling Problem",

abstract = "This paper presents two Reinforcement Learning approaches for the Parallel Machines Job Shop Scheduling Problem. This NP-hard optimization problem where operations have to be assigned to machines of different workcenters, each workcenter having a number of identical parallel machines, is hard to solve and is therefore tackled with learning based methods. The objective used is the minimization of the schedule makespan. Two viewpoints are taken, one where resources are intelligent agents and have to choose what operation to process next, and another where operations themselves are seen as the agents that have to choose their mutual scheduling order. As Reinforcement Learning methods we use a value iteration method (Q-Learning) and a policy iteration method (Learning Automata). The results of both approaches improve on recent published results from the literature.",

keywords = "Reinforcement Learning, Job Shop Scheduling, Parallel Machines Job Shop Scheduling",

author = "{Martinez Jimenez}, Yailen and Wauters Tony and Ann Nowe and Katja Verbeeck and Patrick, {De Causmaecker} and Juliett Suarez and Rafael Bello",

year = "2010",

month = feb,

language = "English",

isbn = "978-959-250-574-2",

series = "Proceedings of the Cuba-Flanders Workshop on Machine Learning and Knowledge Discovery",

booktitle = "Proceedings of the Cuba-Flanders Workshop on Machine Learning and Knowledge Discovery",

}

Reinforcement Learning approaches for the Parallel Machines Job Shop Scheduling Problem. / Martinez Jimenez, Yailen; Tony, Wauters; Nowe, Ann; Verbeeck, Katja; Patrick, De Causmaecker; Suarez, Juliett; Bello, Rafael.

Proceedings of the Cuba-Flanders Workshop on Machine Learning and Knowledge Discovery. 2010. (Proceedings of the Cuba-Flanders Workshop on Machine Learning and Knowledge Discovery).

Research output: Chapter in Book/Report/Conference proceeding › Conference paper

TY - GEN

T1 - Reinforcement Learning approaches for the Parallel Machines Job Shop Scheduling Problem

AU - Martinez Jimenez, Yailen

AU - Tony, Wauters

AU - Nowe, Ann

AU - Verbeeck, Katja

AU - Patrick, De Causmaecker

AU - Suarez, Juliett

AU - Bello, Rafael

PY - 2010/2

Y1 - 2010/2

N2 - This paper presents two Reinforcement Learning approaches for the Parallel Machines Job Shop Scheduling Problem. This NP-hard optimization problem where operations have to be assigned to machines of different workcenters, each workcenter having a number of identical parallel machines, is hard to solve and is therefore tackled with learning based methods. The objective used is the minimization of the schedule makespan. Two viewpoints are taken, one where resources are intelligent agents and have to choose what operation to process next, and another where operations themselves are seen as the agents that have to choose their mutual scheduling order. As Reinforcement Learning methods we use a value iteration method (Q-Learning) and a policy iteration method (Learning Automata). The results of both approaches improve on recent published results from the literature.

AB - This paper presents two Reinforcement Learning approaches for the Parallel Machines Job Shop Scheduling Problem. This NP-hard optimization problem where operations have to be assigned to machines of different workcenters, each workcenter having a number of identical parallel machines, is hard to solve and is therefore tackled with learning based methods. The objective used is the minimization of the schedule makespan. Two viewpoints are taken, one where resources are intelligent agents and have to choose what operation to process next, and another where operations themselves are seen as the agents that have to choose their mutual scheduling order. As Reinforcement Learning methods we use a value iteration method (Q-Learning) and a policy iteration method (Learning Automata). The results of both approaches improve on recent published results from the literature.

KW - Reinforcement Learning

KW - Job Shop Scheduling

KW - Parallel Machines Job Shop Scheduling

M3 - Conference paper

SN - 978-959-250-574-2

T3 - Proceedings of the Cuba-Flanders Workshop on Machine Learning and Knowledge Discovery

BT - Proceedings of the Cuba-Flanders Workshop on Machine Learning and Knowledge Discovery

ER -

Reinforcement Learning approaches for the Parallel Machines Job Shop Scheduling Problem

Abstract

Publication series

Keywords

Fingerprint

Cite this