Robust Design: Seeking the Best of All Possible
Susan M. Sanchez (Naval Postgraduate School)
We describe a framework for analyzing simulation output in order to find solutions that will work well after implementation. We show how the use of a loss function that incorporates both system mean and system variability can be used to efficiently and effectively carry out system optimization and improvement efforts. For models whose behavior depends on quantitative factors, we illustrate how robust design can be accomplished by using simple experimental designs in conjunction with response-surface metamodels. The results can yield new insights into system behavior, and may lead to recommended system configurations that differ substantially from those selected by analysis solely on the basis of mean response. We assume a knowledge base at the level of Chapter 12 of Simulation Modeling and Analysis (Law and Kelton, 2000) but will review essential elements and distribute illustrative examples at the session.
Developing Industrial Strength Simulation Models
Using Visual Basic for Applications (VBA)
Marvin S. Seppanen (Productive Systems)
Since 1984 the author has developed simulation models that use input data from spreadsheets. These original applications used a standalone Basic program to convert Lotus 123® data into Siman Experiment Frames. While this process has evolved overtime, it did not reach a truly viable level until Arena® 3.0 introduced Visual Basic® for Applications (VBA) by Microsoft®. This advanced tutorial demonstrates the basic concepts developed by the author to transfer data between Excel® and Arena. The same techniques can be used to communicate simulation data with a wide range of VBA supported tools, such as Access®, AutoCAD®, and Visio®. Arena permits the model developer to use VBA as the model file is loaded, executed, or terminated or as entities flow through the Arena model modules. This tutorial focuses on the design of Excel workbooks for simulation applications and the transfer of data to/from Arena using VBA.
Groupware and the Simulation
Simon J.E. Taylor (Brunel University)
This paper recognises that good communication and interaction are key factors to the success of a simulation project and suggests that groupware technology can increase the chances of success. To underline this, the paper reviews the process of simulation to illustrate the amount of communication and interaction that must take place during a simulation project. The paper then discusses computer supported cooperative work and groupware, a research field and information technology that has successfully supported communication and interaction in other industries. To illustrate how groupware may by used by the simulation consultant, net-conferencing, exemplified by Microsoft's NetMeeting, is presented. The paper ends with some observations on the future of these applications in simulation modelling.
Inside Discrete-Event Simulation Software: How It
Works and Why It Matters
Thomas J. Schriber (The University of Michigan) and Daniel T. Brunner (Systemflow Simulations, Inc.)
This paper provides simulation practitioners and consumers with a grounding in how discrete-event simulation software works. Topics include discrete-event systems; entities, resources, control elements and operations; simulation runs; entity states; entity lists; and entity-list management. The implementation of these generic ideas in AutoMod, SLX, and Extend is described. The paper concludes with several examples of “why it matters” for modelers to know how their simulation software works, including coverage of SIMAN (Arena), ProModel and GPSS/H as well as the other three tools.
Output Analysis for Simulations
Christos Alexopoulos (Georgia Institute of Technology) and Andrew F. Seila (University of Georgia)
This paper reviews statistical methods for analyzing output data from computer simulations of single systems. In particular, it focuses on the estimation of steady-state system parameters. The estimation techniques include the replication/deletion approach, the regenerative method, the batch means method, and the standardized time series method.
Bayesian Methods for Simulation
Stephen E. Chick (The University of Michigan)
This tutorial describes some ways that Bayesian methods address problems that arise during simulation studies. This includes quantifying uncertainty about input distributions and parameters, sensitivity analysis, and the selection of the best of several simulated alternatives. Focus is on illustrating the main ideas and their relevance to practical problems. Numerous citations for both introductory and more advanced material provide a launching pad into the Bayesian literature.
A Survey of Simulation Optimization Techniques and
James R. Swisher (Mary Washington Hospital), Paul D. Hyden (Cornell University), Sheldon H. Jacobson (University of Illinois at Urbana-Champaign) and Lee W. Schruben (University of California (Berkeley))
Discrete-event simulation optimization is a problem of significant interest to practitioners interested in extracting useful information about an actual (or yet to be designed) system that can be modeled using discrete-event simulation. This paper presents a brief survey of the literature on discrete-event simulation optimization over the past decade (1988 to the present). Swisher et al. (2000) provides a more comprehensive review of this topic while Jacobson and Schruben (1989) covers the literature preceding 1988. Optimization of both discrete and continuous input parameters are examined herein. The continuous input parameter case is separated into gradient and non-gradient based optimization procedures. The discrete input parameter case differentiates techniques appropriate for small and for large numbers of feasible input parameter values.
A Framework for Response Surface Methodology for
H. Gonda Neddermeijer, Gerrit J. van Oortmarssen, Nanda Piersma, and Rommert Dekker (Erasmus University Rotterdam)
We develop a framework for automated optimization of stochastic simulation models using Response Surface Methodology. The framework is especially intended for simulation models where the calculation of the corresponding stochastic response function is very expensive or time-consuming. Response Surface Methodology is frequently used for the optimization of stochastic simulation models in a non-automated fashion. In scientific applications there is a clear need for a standardized algorithm based on Response Surface Methodology. In addition, an automated algorithm is less time-consuming, since there is no need to interfere in the optimization process. In our framework for automated optimization we describe the many choices that have to be made in constructing such an algorithm.
Mathematics for Simulation
Shane G. Henderson (University of Michigan)
I survey several mathematical techniques and results that are useful in the context of stochastic simulation. The concepts are introduced through the study of a simple model of ambulance operation to ensure clarity, concreteness and cohesion.
On Hybrid Combination of Queueing and Simulation
Nico M. van Dijk (University of Amsterdam / Incontrol Business Engineers)
"Should we pool separate queues into a single queue or not?" A question as practical as for daily-life situations such as at a bank, a hospital or a service center as well as for technical applications such as in manufacturing or telecommunications (multi-plexing). A question that involves fundamental insights of queuing theory. A question that is still open for research. A question that in realistic situations not only benefits from but even requires a hybrid combination of analysis and simulation.
Using Simulation for Option Pricing
John M. Charnes (The University of Kansas)
Monte Carlo simulation is a popular method for pricing financial options and other derivative securities because of the availability of powerful workstations and recent advances in applying the tool. The existence of easy-to-use software makes simulation accessible to many users who would otherwise avoid programming the algorithms necessary to value derivative securities. This paper presents examples of option pricing and variance reduction, and demonstrates their implementation with Crystal Ball 2000, a spreadsheet simulation add-in program.
Creating Distributed Simulation Using DEVS
Bernard P. Zeigler and Hessam S. Sarjoughian (University of Arizona)
We briefly review the theory of modeling and simulation and its support for constructing distributed simulations. Formal representation of simulation models can contribute to a number of aspects in the modeling and simulation enterprise. Separation of models from simulation execution engines is a prerequisite transferring model among phases of a project as well as from project to project. The Discrete Event System Specification (DEVS) formalism, drawing on its system theoretic basis, provides a number of important properties such as hierarchical, modular composition, universality and uniqueness that can support development of simulation models and environments their development. An layered architecture for supporting comprehensive M&S environments is discussed that unifies the theoretical framework with implementation in distributed computational environments.