A Fuzzy-neural Resemblance Approach to Validate Simulation Models

摘要

Validation is one of the most important steps in developing a reliable simulation model. It evaluates whether or not the model forms a representation of the simulated system accurate enough to satisfy the goals of the modelling study. The methods that are currently available for model validation are binary in nature in the sense that they only allow either to accept or reject the validity of the model. In this paper, we develop a new, fuzzy set theoretic method that allows to express degrees of model validity and that is hence continuous in nature. The method employs a fuzzy-neural machine learning algorithm and makes use of a new concept in fuzzy set theory known as resemblance relations. By a computational experiment, we demonstrate how our method can be used to discriminate more from less valid simulation models of a particular manufacturing process.