My SciELO
Services on Demand
Article
Indicators
- Cited by SciELO
Related links
- Similars in SciELO
Share
Revista Cubana de Ciencias Informáticas
On-line version ISSN 2227-1899
Abstract
MENDEZ-HERNANDEZ, Beatriz M et al. Bi-objective approach based in Reinforcement Learning to Job Shop scheduling. Rev cuba cienc informat [online]. 2017, vol.11, n.2, pp. 175-188. ISSN 2227-1899.
ABSTRACT Scheduling problems require organizing the execution of tasks which share a finite set of resources, and these tasks are subject to a set of constrains imposed by different factors. This kind of problems frequently appears in many production and service environments. The problem is to optimize one or more criteria represented by objective functions. In this paper, the main objectives to optimize were analyzed for Job Shop scheduling problems. After that, a bi-objective algorithm was proposed based on the Pareto Front and using Reinforcement Learning, which optimizes two objectives: the makespan and the total flow time, and this algorithm was applied to benchmarks. To finish, successful results of the algorithm are described according to two metrics proposed in the literature.
Keywords : Job Shop; multi-objective; Pareto; Reinforcement Learning.