Genetic algorithms with cluster analysis for production simulation
Robert Entriken, Siegfried Vössner · 1997
This paper describes the application of a Genetic Algorithm to production simulation.The simulation is treated as a detailed, stochastic, multi-modal function that describes a performance statistic.Our aim is to optimize (or at least improve) the performance of the system.In our experiments, we modeled a real-world production line for printed circuit boards that has many products and must often be retooled or reconfigured.Since the product line is always changing, with half of the products turning over within a year, the job of configuring and fine tuning the production line is never ending.Our experiments show that a Genetic Algorithm when attached to the simulation model can provide excellent support for this process.This combination can be used to obtain quick and stable results that do indeed indicate the direction to improved production.