Date of Award


Document Type


Degree Name

Master of Science


Department of Operational Sciences

First Advisor

Edward A. Pohl, PhD

Second Advisor

W. Paul Murdock, PhD


The overall objective of this research effort was to formulate a preventive maintenance strategy for AMRAAM missiles subject to extended captive carry flight time. A preventive maintenance policy is only applicable if the item in question is aging, or deteriorating with time. Therefore, a supporting objective of this research is to characterize the aging process of the missile system through a non-parametric analysis of its Mean Residual Life (MRL) function. Three non-parametric, censored-data MRL function estimation techniques discussed in the literature are examined via a numerical example. All three estimation techniques provide MRL functions that exhibit greatly exaggerated decreasing trends compared to the MRL function of the underlying distribution in the example. A semi-parametric technique for estimating the MRL function is developed that shows dramatic improvement over the non-parametric results. Although the MRL analysis of the current AMRAAM failure data failed to provide evidence that the missile system is aging, three preventive maintenance policies discussed in the literature are investigated. The traditional approach of preventive maintenance policy optimization via cost function minimization requires the cost of a system failure be explicitly known. However, the penalty for a system failure is often subjective and difficult to express in monetary terms. A "reliability cost model" is developed whereby system reliability for each policy is expressed as a function of cost. This technique allows a direct assessment of the trade-off between cost and projected system reliability. Theoretical results are presented and the performance of the model applied to empirical data is assessed via a Monte Carlo simulation.

AFIT Designator


DTIC Accession Number