Paul J. Sanchez

Very large fractional factorial and central composite designs

We present a concise representation of fractional factorials and an algorithm to quickly generate resolution V designs. The description is based on properties of a complete, orthogonal discrete-valued basis set called Walsh functions. We tabulate two-level resolution V fractional factorial designs, as well as central composite designs allowing estimation of full second-order models, for experiments involving up to 120 factors. The simple algorithm provided can be used to characterize even larger designs, and a fast Walsh transform method quickly generates design matrices from our representation.

Designing simulation experiments: Taguchi methods and response surface metamodels

G. Taguchi (1987) has made an innovative contribution to quality planning activities through the integrated use of loss functions and orthogonal arrays. The authors focus on the improvement and implementation of some of these techniques in the simulation arena. The orthogonal arrays advocated by Taguchi are related to classical experimental designs, which have played important tactical roles in the exploration of mathematical metamodels for the simulation response surface. However, the loss function and the associated robust design philosophy provide fresh insights into the process of optimizing or improving the simulations performance. The authors use examples to illustrate concepts such as the simultaneous treatment of variability and mean of performance measures, strategies for achieving system robustness, and implementation of noise through factorial designs. Relationships to other issues in designing and analyzing simulation experiments, such as response surface metamodels and variance reduction, are discussed.<<ETX>>

Effective Engineering Design through Simulation

Abstract This paper presents a framework for designing, analyzing and improving systems and processes via discrete event simulation. The framework incorporates a robust design philosophy into a response surface metamodeling approach, and the simulation setting provides the analyst with an increased level of control relative to industrial experimentation. System optimization and improvement efforts can be carried out efficiently and effectively, providing insights into system behavior and suggesting optimal system configurations which may yield substantial improvements over those selected using more traditional approaches. One noteworthy benefit of the simulation framework is that robust design methodologies can be applied prospectively — at the inception and conceptualization phases of an engineering design project. We illustrate the method by considering the design of a small job shop.

A model for frequency domain experiments

We present a meta-model which is useful for understanding simulation frequency domain experiments. This model consists of polynomial gain followed by a linear filter with additive noise. The assumptions for performing frequency domain experiments are thus made explicit. We demonstrate how the model leads to a straightforward mechanism for factor screening via statistical hypothesis testing.

Fundamentals of simulation modeling

This publication contains reprint articles for which IEEE does not hold copyright. Full text is not available on IEEE Xplore for these articles.

As simple as possible, but no simpler: a gentle introduction to simulation modeling

We start with basic terminology and concepts of modeling, and decompose the art of modeling as a process. This overview of the process helps clarify when we should or should not use simulation models. We discuss some common missteps made by many inexperienced modelers, and propose a concrete approach for avoiding those mistakes. After a quick review of event graphs, which are a very straightforward notation for discrete event systems, we illustrate how an event graph can be translated quite directly to a computer program with the aid of a surprisingly simple library. The resulting programs are easy to implement and computationally are extremely efficient. The first half of the paper focuses principles of modeling, and should be of general interest. The second half will be of interest to students, teachers, and readers who wish to know how simulation models work and how to implement them from the ground up.

Simple movement and detection in discrete event simulation

Many scenarios involving simulation require modeling movement and sensing. Traditionally, this has been done in a time-stepped manner, often because of a mistaken belief that using a pure discrete event approach is infeasible. This paper discusses how simple motion (linear, uniform, two-dimensional) and simple sensing can be modeled with a pure discrete event approach. We demonstrate that this approach is not only feasible, it is often more desirable from several standpoints.

Design of frequency-domain experiments for discrete-valued factors

We provide theoretical justification for the use of an existing technique for studying binary-valued inputs in the Fourier frequency domain. Extending this work, we propose two methods for handling general discrete-valued factors, and illustrate their use. The two methods are found to have complementary applicability. When combined with previously developed techniques, they generalize the frequency- domain approach to encompass discrete as well as continuous inputs.

Simulating pirate behavior to exploit environmental information

Recent years have seen an upsurge in piracy, particularly off the Horn of Africa. Piracy differs from other asymmetric threats, such as terrorism, in that it is economically motivated. Pirates operating off East Africa have threatened maritime safety and cost commercial shipping billions of dollars paid in ransom. Piracy in this region is conducted from small boats which can only survive for a few days away from their base of operations, have limited survival in severe weather, and cannot perform boarding operations in high wind or sea state conditions. In this study we use agent models and statistical design of experiments to gain insight into how meteorological and oceanographic forecasts can used to dynamically predict relative risks for commercial shipping.

Variance reallocation in Taguchi's robust design framework

The appropriate use of antithetic random variates has been shown to improve the precision of response surface model estimation for simulation. We apply this approach to simulation experiments designed to determine operating conditions that reduce response variability by using Genichi Taguchi’s parameter design framework. Antithetic random number streams can be viewed as another level of complexity in the experiment design: we call this class of simulationspecific factors artificial factors. A simple example illustrates how antithetic random variates may be beneficial for robust design in simulation settings.