Date of Award

3-16-2009

Document Type

Thesis

Degree Name

Master of Science in Operations Research

Department

Department of Operational Sciences

First Advisor

Shane N. Hall, PhD

Second Advisor

James T. Moore, PhD

Abstract

This thesis describes a method that optimally deploys weather sensors of all types in a battlefield environment. Gridded climatology models are used to determine an estimate for the weighted frequency of occurrence of operationally significant inclement weather events. That data is used to formulate a series of preemptive Binary Integer Linear Programs that maximize detection of expected operationally significant inclement weather occurrences within the constraints of feasibility of sensor deployment, sensor operational lifespan and the sensor’s ability to detect the operationally significant inclement weather elements. The preemptive Binary Integer Linear Programs are combined into a single objective function that maintains the preemptive nature of the original objective functions. The BILP solutions are described as a meteorology and oceanographic collection plan supporting a particular military campaign. A method for sensitivity analysis of differing BILP optimal solutions is provided. Various realistic instances of the problem are solved to optimality and analyzed to demonstrate that the problem formulation accurately captures all aspects of the problem. This type of analysis was not possible before this methodology was developed.

AFIT Designator

AFIT-GOR-ENS-09-06

DTIC Accession Number

ADA500327

Comments

Co-advised thesis.

Included in

Meteorology Commons

Share

COinS