An Evolutionary Neural System for incorporating Human Expert Knowledge into the UA-FLP

Áreas de investigación:

• Methodology - Machine learning

Año:

2014

Tipo de publicación:

Artículo

Palabras clave:

Evolutionary Computation, Artificial Neural Networks, Unequal Area Facility Layout Problem, Layout Representation, Basis functions

Autores:

• García-Hernández, L.
• Pérez-Ortiz, María
• Araúzo-Azofra, A.
• Salas-Morera, L.
• Hervás-Martínez, César

Journal:

Neurocomputing

Volumen:

135

Páginas:

69-78

Mes:

July

ISSN:

0925-2312

Nota:

JCR(2014): 2.083 Position: 36/124 (Q1) Category: COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Abstract:

This paper presents an hybrid system for incorporating human expert knowledge into the Unequal Area Facility Layout Problem. For that matter, a subset of facility designs are generated (using a Genetic Algorithm) and then, evaluated by an human expert. The hybrid system is giving a mark, with the aim of substituting the human expert knowledge, avoiding his/her fatigue and burden. The novel proposed approach was tested under a real case study made of 365 facility layout designs from an ovine slaughterhouse. The validation phase of the intelligent model presented is performed by means of a new subset of 181 facility layout designs with the evaluation provided by a different human expert. The results of the experiment, which validate the performance of the proposed approach, are presented and discussed in this study.

Comentarios:

JCR(2014): 2.083 Position: 36/124 (Q1) Category: COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE