Date of Award
Master of Science
Department of Operational Sciences
Brian J. Lunday, PhD
The 96th Test Wing at Eglin Air Force Base manually schedules a fleet of approximately 26 aircraft to conduct a range of missions over a one-to-two year planning period. This study automates the scheduling process, does so in a manner that optimizes multiple planning goals related to aircraft availability for training, and provides the 96th Test Wing with a software tool for the implementation that can be used by operational analysts within the command. We formulate the scheduling problem as a multiobjective, nonlinear, binary integer math program that seeks to maximize both the lowest percent of time any aircraft is available for training and the lowest percent of aircraft available for training for any week. Applying the Weighted Sum Method for multiobjective optimization, a conversion of nonlinear operations yields a binary integer program that is directly solvable via a commercial solver. An examination of the multi-objective nature of the problem identified a lack of tension between the objectives, so empirical testing affixes equal weights to the well-scaled objective function components. The 96th Test Wing directed the use of Excel as a modeling environment and What's Best! by Lindo Systems, Inc. for their analysts to use. Subsequent testing examined for increasing time horizons the ability of the model and solver combination to develop optimal or near-optimal schedules for increasing levels of mission densities. The required computational effort was not predictable by mission density level, but increasing levels did yield instances not solvable to optimality within four hours for each time horizon. Optimal solutions can be readily identified within a few seconds for the current fleet and current or higher mission densities for 8-, 26-, and 52-week schedules, using weekly granularity for mission scheduling.
DTIC Accession Number
Hoops, Sarah E., "Optimal Scheduling Of Aircraft Test and Evaluation Fleets to Balance Availability for Testing and Training" (2022). Theses and Dissertations. 6298.