Accession Number : ADA185247

Title :   Automating Response Surface Methodology for the Arsenal Exchange Model.

Descriptive Note : Master's thesis,

Corporate Author : AIR FORCE INST OF TECH WRIGHT-PATTERSON AFB OH SCHOOL OF ENGINEERING

Personal Author(s) : Yielding, Gary E

PDF Url : ADA185247

Report Date : May 1987

Pagination or Media Count : 307

Abstract : Response Surface Methodology (RSM) is automated for the Arsenal Exchange Model. The purpose is to give the using organization (AF/XOXR-MAA) the capability of rapidly answering 'what-if' questions in real time on force structure problems. Currently, the AEM requires approximately three hous CPU time for one execution. By using RSM to build a replacement model for the AEM, a second-order polynomial equation for this thesis, the execution time is brought to within seconds. So, the RSM technique builds a replacement model for the AEM, given a specific scenario, that provides predictability within seconds. Also, using experimental designs that are othogonal and rotatable, the replacement model (second-order polynomial equation) can be used very effectively and efficiently for informational analysis (same as sensitivity analysis). Much more information can be garnered from the replacement model than the real model (the AEM). So, besides the advantage of predictability, the RSMed built replacement model provides much more informational analysis than what could have been garnered by applying sensitivity analysis to the AEM.

Descriptors :   *OPERATIONS RESEARCH, *RESOURCE MANAGEMENT, *AIR FORCE OPERATIONS, ALLOCATIONS, STRATEGIC WEAPONS, MILITARY FORCES(UNITED STATES), CODING, AUTOMATION, APPROXIMATION(MATHEMATICS), DIFFERENTIAL EQUATIONS, MISSION PROFILES, TARGETING, MATHEMATICAL PREDICTION, SENSITIVITY, THESES

Subject Categories : Operations Research
      Computer Programming and Software

Distribution Statement : APPROVED FOR PUBLIC RELEASE