SIMULATION: Transactions of The Society for Modeling and Simulation International

Wildberger, A. Martin

doi: 10.1177/003754979005500301pmid: N/A

Stockton, Chip

doi: 10.1177/003754979005500302pmid: N/A

McInnis, Michael L.; Quigley, Thomas M.; Vavra, Martin; Reed Sanderson, H.

doi: 10.1177/003754979005500303pmid: N/A

A simple deterministic model was developed to predict animal unit months (AUM's) and live weight gains of beef cattle grazing specific range types in eastern Oregon. The model can provide values for 3 levels of spatial resolution (pasture, mapping unit, and slope/ proximity to water "cells" within map ping units) and 5 monthly periods from May 15 to October 14. Operation of the model begins by calculating forage availability as a factor of forage biomass, usable acres, and desired forage utiliza tion. Grazing capacity (AUM's) is calculated by comparing forage availabil ity with the dry matter forage requirement of a 1,000 lb animal unit for 30 days. Live weight gains are calculated by comparing forage availability to dry matter forage intake, crude protein intake, and digestible energy intake of yearling heifers. The model can be used in planning range improvements and coordinating livestock management with other rangeland activities.

Raczynski, Stanislaw

doi: 10.1177/003754979005500304pmid: N/A

The Queuing Model Generator (QMG), described in this article, is a module of the environment of the simulation language PASION. It contains a block-diagram editor which permits the user to define the structure of the model, and a program generator which generates the corresponding PASION code. This code is automatically translated to PASCAL. QMG is transpar ent, (i.e. the languages PASION and PASCAL are not visible for "non-program mer" users.) However, if the user knows PASION or PASCAL, then he can work with the resulting code, creating more complex models. An application in flexible manufacturing systems design is described.

Hood, Stephen T.; Mason, Keith P.; Mildren, Wayne J.

doi: 10.1177/003754979005500305pmid: N/A

A computer simulation of a receiver intercepting radar emissions in a tactical maritime environment has been developed as a means of stimulating a knowledge-based system for signal interpretation. The development of the model from concept, through a prototype in Prolog, to the current version in C is described in this paper. The model is designed to run in a multi processing environment providing real-time simulation of the signal environment to allow the performance of the knowledge- based system to be evaluated. The model provides a language for specifying scenarios in an object-oriented style, and uses inter process communication facilities provided by the UNIX operating system to connect to the application system. Interactive tools to control the runtime characteristics of the simulation are also provided.

Banks, J.; Carson, J.S.; Sy, J.N.

doi: 10.1177/003754979005500306pmid: N/A

The Red Book lies quietly on my shelf, its cover frayed through years of heavy use, some pages torn.The book and I were both young once. Before 1974 when Tom Schriber was still pregnant with the ideas for the book, I employed a pre-publication version to equip bright and eager students with the skills to practically apply this innovative problem-solving tool, GPSS.As the years went by, however, the book and I both became grayer. Updated versions of GPSS appeared on various computer platforms. Microcomputers became the tool of choice and friendlier versions of GPSS became available. Through it all, the Red Book never changed, although its price always rose. Rumor had it that Tom Schriber wanted to revise and expand the book so we waited and remained loyal, but nothing happened. We were very patient, but now it is 1990.

Seila, Andrew F.; Banks, Jerry

doi: 10.1177/003754979005500307pmid: N/A

For many financial models implemented in electronic spreadsheets, input data values frequently are random variables because they are actually estimates of unknown quantities. As a result, the bottom-line performance measure of the model is a random variable, and risk is associated with decisions based upon it due to the uncer tainty in its value. We describe in detail how to evaluate this risk using simulation in a spreadsheet and illustrate the procedure with an example. Formulas for generating random variates from many common distributions using LOTUS 1-2-3 are given, and data analysis considerations are also discussed.

Unger, Brian

doi: 10.1177/003754979005500308pmid: N/A

Stockton, C.G.

doi: 10.1177/003754979005500309pmid: N/A

Le Minh, Do; Sherif, Y.S.

doi: 10.1177/003754979005500310pmid: N/A

This paper presents a simple method to obtain a strictly positive lower confidence limit for the mean of a positive random variable. Although this method involves some approxi mations, it is shown empiri cally that it works well for a large number of cases.