MIL-HDBK-472 Maintainability Prediction
Maintainability is the measure of the a system or device to be retained in or restored to specific conditions when maintenance is performed by personnel having specified skill levels, using prescribed procedures and resources at each prescribed level of maintenance and repair.
The prediction of the expected number of hours that a system or device will be in an inoperative or "down" while it is undergoing maintenance is of vital importance because of the adverse effect that excessive downtime has on mission success. A technique must be utilized to predict its maintainability as early as possible during the design phase. This prediction should be updated continuously as the design progresses.
A maintainability prediction procedure highlights those areas of poor maintainability, which justify product improvement, modification, or a change of design. It permits early assessment of whether the predicted downtime, the quality, quantity of personnel, tools and test equipment are adequate and consistent with the needs of the system operational requirements.
Each maintainability prediction procedure included in MIL-HDBK-472 depends upon reliability and maintainability data and experience. The techniques contained in MIL-HDBK will fit most applications, but they have not been tested for generality or for consistency from one to another.
Each maintainability prediction technique utilizes procedures that are specifically designed to satisfy its method of application. However, all maintainability prediction methods are dependent upon at least two basic parameters:
There are many sources that record the failure rate of parts as a function of use and environment. This failure rate is expressed as the number of failures per unit of time. A typical measure is "Failures Per Million Hours."
Corrective maintenance actions consist of the following tasks:
The time to perform each of these tasks is an element of MTTR. Hence the task times are called MTTR elements.
Other information is necessary too: