ABSTRACT
MTBFs are calculated and estimated, but testing to demonstrate the MTBF is still somewhat of a mystery. Enlightened companies where components are burned-in before assembly and assemblies are power-aged to eliminate infant mortality, often do not look for the added profits associated with the measurement of MTBF, which can bring product improvement. It is almost as easy to measure compliance to an estimated or calculated MTBF at the end of a production line as it is to measure compliance to an AQL at receiving inspection. In general, life tests with more UTHs in the test, and higher accept numbers (c) are more discriminatory than those with fewer hours and lower acceptance numbers. The best policy is to evaluate the risks involved by analyzing the OC curves to optimize the risks of the life test plan for use.
