An independent appraisal of the polaris operational testing procedures.

Abstract

The first section of this paper is concerned with determining the required lower confidence limit that must be met by testing after a missile system becomes operational. Some of the costs of making decisions about the required system reliability lower confidence limit are discussed. Two possible cost effective models for determining the optimum test size are suggested. The second section of this paper is concerned with the effects of changing the number of missiles tested in each year while maintaining the total number of systems tests constant over the missile system's projected operating life. In other words, in this section, the effects of testing heavily in the first years versus testing heavily at the end of the system's life or versus testing uniformly throughout the life of the system are compared. For this comparison, the sum of the variances of the estimators determined from the results of the tests conducted in each year is obtained. This value is compared for six representative distributions of testing throughout an estimated system life of ten years.