Resumen

CASTRILLON, Omar Danilo; SARACHE, William Ariel  y  HERRERA, Santiago Ruiz. A Two-Class Bayesian Classifier to Select the Best Priority Rule in a Job Shop: Open Shop Scheduling Problem. Rev.EIA.Esc.Ing.Antioq [online]. 2019, vol.16, n.31, pp.57-64. ISSN 1794-1237.  https://doi.org/10.24050/reia.v16i31.867.

The aim of the present paper to select, through a two-class Bayesian classifier, the best priority rule to solve a Job Shop: Open Shop scheduling problem. In a first phase, the design of the classifier, trained with 300 randomly problems, is exposed. In 150 of them, the best priority rule for sequencing was FIFO (First In First Out) while in the rest was the LPT (Long Process Time). In a second phase, a set of 300 different problems, with the same characteristics of the first phase, were randomly generated. These problems were classified previously (without sequencing them) through the proposed Bayesian technique. The results show that, in 96% of cases, the proposed classifier identifies the best priority rule to sequence the orders.

Palabras clave : Production scheduling; Priority Rules; Bayesian classifier; Job Shop: Open Shop.

        · resumen en Español | Portugués     · texto en Español     · Español ( pdf )