Interpretation summary for Estimation Test Plan

Use Estimation Test Plan to determine the sample size necessary to estimate a percentile or reliability with a specified amount of precision.

Estimation test plans are similar to classical sample-size problems, but computations are more intensive because the data are usually censored. Estimation test plans allow for right- or interval-censored data.

Data description

You are developing a new type of insulation and want to determine the sample sizes necessary to estimate the 10^th percentile with the distance from the lower bound to the estimate to be within 100, 200, and 300 hours. To do this estimation, you will run tests on small specimens for 1000 hours.

You have the following planning information:

You expect approximately 12% of the specimens to fail in the first 500 hours of the test.
You expect approximately 20% of the specimens to fail by the end of 1000 hours.
Insulation failure times follow a Weibull distribution.