Reliability Prediction for One-Shot Devices
A one-shot device is defined as an item that is required to perform its function once during normal operational use. In general, such items will be destroyed during their normal operation (e.g., motors, fuzes, warheads, etc.) and will therefore be no-test items. Also, one-shot devices are required to operate for only a relatively short time.
For these reasons, it can be assumed for prediction purposes that failure of any one-shot device is independent of time and can be expressed as a fixed probability of occurrence, (PF). Reliability (i.e., probability of successful operation) is then given by (I-PF ). Data for various one-shot devices is given in Data Tables.
When using the prediction methods described in previous paragraphs, one-shot devices must be identified separately, and their reliabilities derived from the data given in Appendix A. These reliabilities should then be combined with the reliabilities of other items in the design, according to the system reliability model.
If data is not given in Table A-59, estimates of reliability should be made by comparison with similar items, and these values used until more specific data becomes available, (e.g., from testing programmes).
 |
The data given in Appendix A is related to a temperate climate. The possible effects of more severe environmental conditions should be assessed, as necessary, by the analyst, using the best available information. The sources of such information, and any alternative data used in the prediction, should be recorded.
|
The reliability required from one-shot devices is generally high. It is most important, therefore, that their design and manufacturing processes are evaluated in as much depth as possible, using both quantitative and qualitative analysis techniques. These are described more fully in Reference 1, which is described in the Bibliography.