Measuring Missile Reliability In Prelaunch Environments

This paper examines some of the problems involved in measuring missile reliability during its prelaunch phase. This is an important problem for three reasons: 1) a missile typically is exposed to many kinds of operating environments over long periods of time before it is launched; 2) a missile's reliability history in its prelaunch operating environments is an index of its in-flight reliability; and 3) the major cost of a missile weapon system derives from the consequences of its prelaunch reliability behavior. Most of the emphasis to date in reliability measurement efforts has been focused, on the inflight environment, to the near-exclusion of the important prelaunch environment. Attempts to infer prelaunch reliability on the basis of data collected for other purposes, such as supply transactions, have shown that this is not an adequate substitute for direct measurements. This document provides an approach to the problem specifically oriented to the collection and analysis of reliability data. Reliability measurement requirements stem directly from the basic definition of reliability: the probability that an aggregate will successfully perform a specified task in a specified environment over a specified period of time. Several of the most important prelaunch operating environments are: turn-on (-off), checkouts, countdowns, operating alert, standby alert, handling, transportation, periodic maintenance and storage.