Simulation metamodelling with neural networks: an experimental investigation

Date
2002
Authors
Sabuncuoğlu, İ.
Touhami, S.
Advisor
Supervisor
Co-Advisor
Co-Supervisor
Instructor
Source Title
International Journal of Production Research
Print ISSN
0020-7543
Electronic ISSN
Publisher
Volume
40
Issue
11
Pages
2483 - 2505
Language
English
Type
Article
Journal Title
Journal ISSN
Volume Title
Series
Abstract

Artificial neural networks are often proposed as an alternative approach for formalizing various quantitative and qualitative aspects of complex systems. This paper examines the robustness of using neural networks as a simulation metamodel to estimate manufacturing system performances. Simulation models of a job shop system are developed for various configurations to train neural network metamodels. Extensive computational tests are carried out with the proposed models at various factor levels (study horizon, system load, initial system status, stochasticity, system size and error assessment methods) to see the metamodel accuracy. The results indicate that simulation metamodels with neural networks can be effectively used to estimate the system performances.

Course
Other identifiers
Book Title
Keywords
Computer simulation, Large scale systems, Mathematical models, Neural networks, Problem solving, Robustness (control systems), Job-shop systems, Production engineering
Citation
Published Version (Please cite this version)