What is the purpose of a reliability model?

A reliability model presents a clear picture of functional interdependencies and provides a framework for developing quantitative system level reliability estimates to guide in the design tradeoff process.  Models are helpful for the following reasons:

1.  Identification of single points of failure;

2. Making numerical allocations;

3. Evaluating complex redundant configurations;

4. Showing all series-parallel relationships; and

5. Allowing summarization of all factors effecting system reliability.

System models are derived from, and traceable to, functional requirements.  Models may take inputs from part level reliability predictions, test data, field data, customer requirements and use profiles.  The modeling process often starts at the device level by determining failure rates for components (e.g., published values, vendor test data, field data, etc.).  Component failure rates may be adjusted to take into account different electrical, thermal, and mechanical stresses and are then summed to determine higher assembly level failure rates.  This simple summation is possible because studies have shown that the failure rate for electronic equipment is relatively constant during the useful life of an equipment.  Models vary from relatively simple to highly detailed, taking into account things such as duty cycles, service life limitations, wear out items, varying environments and dormant conditions.  The scope of the model usually depends on the type and amount of information available for use and the criticality of the system under consideration.

For more Information: