Abstract :
The problem of "storage" of electronic parts and equipment first came into focus with the publication of the AGREE Committee Report on Task 8 in 1957. At that time, this problem was recognized as having important logistics connotations. For logistics planning, it was necessary to know what would be the failure rate of components and equipments when they were "on-the-shelf" for a long interval of time. Only general conclusions relating to gross equipments were given in that report. Since that time, most of the attention has been given to the development of operating failure rates, but with the development of new and more sophisticated systems, this problem of storage began to take on new dimensions and new significance. Storage, in addition to its meaning of "on-the-shelf" also had to be extended to include those components which were effectively "stored" within a piece of equipment. These are components which are installed as part of an operating system, but not subject to electrical stress and consequently are electrically inert. This is a key consideration - some of these components turn out to be highly critical - they aré inert but when required to operate they must operate or the mission could be aborted. It is for this reason that the storage problem becomes important. Since the Air Force has imposed the requirements of AFBM Exhibit 58-10 dated i June 1959 on Weapon System contractors, the behavior of equipments under storage conditions in this broad sense becomes a contractual requirement. Paragraph 4.7 of this Exhibit is clear when it states: "The contractor shall conduct studies and tests to determine the effects of storage conditions on the reliability of his product etc." To satisfy this requirement a study was made to determine whether storage failure rates were available in the published literature. Specific quantized failure rates have rarely been determined or published, but some limited material for specific parts has appeared which can be used as a basis for the establishment of storage failure rates within magnitude boundries, i.e., 100 times or 1000 times better than active. The details of these source investigations and the method of performing the subsequent analyses are reported within the paper. Data for this study was - obtained from Batelle Memorial Institute, Burroughs Co., RCA, The Martin Company, the AGREE report and the New York Naval Shipyard Materials Laboratory. A short bibliography listing additional qualitative information is also given for further investigations and study.
