Potential of Reduced Experimental Designs in Object-Oriented Modeling of a Manufacturing Department for Film Production
From Proceeding of European Simulation Multiconference,ESS94, Barcelona, June 1-3 1994
Agostino G.Bruzzone

---

INTRODUCTION

Modular simulation tools are increasingly required by the modern industrial world. Along with many other advantages, this approach can be used to obtain a much more accurate design of the specifications for the high-level blocks through the experience acquired on the base application.
The simulation models produced with object-oriented tools are particularly suitable for these improvements. In fact, object logic can be used to define the base elements on which to operate and therefore, given the modularity of these components, upgrade are very simple to perform.
Within the applications developed in close collaboration with the industrial world, this becomes an extremely important factor. In this type of situation, the products requested are normally designed for continuous expansions and updates and whoever has this type of experience is very familiar with the great need for new potentials required by industrial users from simulators as such devices are used.
This means that the number of variables involved often continues to expand until becoming very high and therefore involving extended experimental campaigns. This fact is also matchedn by an increase in running time due to the greater accuracy of the model.
Thus, it is very important to carry out an accurate evaluation of the information that can be obtained using a reduced number of experiments.
This study presents a typical object-oriented industrial simulation case in which the work generated has gradually expanded to considers aspects relative to the real plant.
The final objective is to present a quantity-based evaluation of the affect of the variables considered which are independent from the plant situation in question. Therefore, reduced experimental designs were carried out to identify the relationships between the different factors.
The results of these reduced designs, having obtained a statistical validation and verified their mutual compatibility, were summarized in a single meta-model used for the general analysis.

Return to the Title