Demonstration test plans are of two types: substantiation tests and reliability tests.
For a substantiation test, the hypotheses are as follows.
- H0: The redesigned system is no different than the current system.
- H1: The redesigned system is better than the current system.
For a reliability test, the hypotheses are as follows.
- H0: The system reliability is equal to a specified value.
- H1: The system reliability is greater than a specified value.
The hypotheses can be defined in terms of the following values:
- Scale (Weibull or exponential distribution) or location (other distributions)
- H0: Location (Scale for Weibull/Exponential) = T
- H1: Location (Scale for Weibull/Exponential) > T
- H0: Pth percentile = T
- H1: Pth percentile > T
- Reliability at a particular time
- H0: Reliability at time T = p
- H1: Reliability at time T > p
- Mean time to failure (MTTF)
- H0: MTTF = T
- H1: MTTF > T
If you pass the demonstration test (the number of failures ≤ M), then you reject H0 at the confidence level that you selected for the demonstration test plan (default is 95%).
If you fail the demonstration test (the number of failures > M), then you fail to reject H0. M is the maximum number of failures allowed according to the demonstration test plan.