University of Oulu

Production optimization on PCB assembly lines using discrete-event simulation

Saved in:
Author: Gebus, Sébastien1; Martin, Olivier1; Soulas, Alexandre1;
Organizations: 1University of Oulu, Faculty of Technology, Control Engineering Laboratory
Format: ebook
Version: published version
Access: open
Online Access: PDF Full Text (PDF, 0.8 MB)
Persistent link:
Language: English
Published: 2004
Publish Date: 2004-09-21


This report describes how discrete-event simulation can be used in production optimisation of electronics assembly lines. Currently, many decisions concerning production are based on workers experience. However, understanding of the parameters influencing production is a challenging task, especially for lines with a wide variety of products. Operation could be improved by analysing bottlenecks and their impact on overall line capacity. Nowadays there are various new tools to extend traditional quality and process-time techniques such as flowcharts or spreadsheets to manage production.

This work focused on how discrete-event simulation could be used in comparing production alternatives to improve production in electronics manufacturing by providing a better understanding of the production environment. It can be used in a straightforward way to test different scenarios for improvement or it can provide information for designing new production facilities. This ability to simulate a real system’s behaviour according to some predefined parameters can also be used as an evaluation tool for new control methods. The present simulator can be used as a platform in comparing optimisation and scheduling approaches before implementation.

An experimental framework for electronics manufacturing plant is described, and modelling choices are focused on comparisons of scheduling policies. PKC Group plant producing control cards for the telecommunication was used for the selection and testing of modelling alternatives. Good results were obtained for basic scheduling policies, allowing comparison between different options. The model could now be used to develop intelligent optimisation methods through links with external software. Among others, genetic algorithms have been considered as a possible choice. Any heuristically based method would benefit greatly of the ability that discrete-event simulation has to mimic real processes.

see all

Series: Control Engineering Laboratory. Report A
ISBN: 951-42-7517-9
ISBN Print: 951-42-7372-9
Issue: 24
Copyright information: © University of Oulu, 2004. This publication is copyrighted. You may download, display and print it for your own personal use. Commercial use is prohibited.