Operational Testing

Background
Operational testing is basically repairable systems analysis using the Crow Extended model. The general assumptions associated with the Crow Extended model do not change. However, in operational testing, the calculations are conducted with the assumption that $$\beta = 1\,\!$$. Therefore, only delayed fixes, BD modes (test-find-test), are allowed. BC modes and fixes during the test cannot be entered. In this scenario, you want a stable system such that the estimate of $$\beta\,\!$$is close to one. The $$\beta = 1\,\!$$ assumption can be verified by checking the confidence bounds on $$\beta\,\!$$. If the confidence bounds include one then you can fail to reject the hypothesis that $$\beta = 1\,\!$$.

Under operational testing:
 * It is not a developmental test since the final product has been fielded.
 * Delayed fixes (BD modes) only.
 * Configuration is fixed and design changes are kept to a minimum.
 * Focus on age-dependent reliability.
 * Testing is generally conducted prior to full production.

Operational testing analysis could also be applied to a system that is already in production and being used by customers in the field. In this case, you will be able to verify the improvement in the system's MTBF based on the specified delayed fixes. Based on this information, along with the cost and time to implement, you can determine if it is cost-effective to apply the fixes to the fielded systems.