Select how to inspect the data for failures. In data with right censoring, the test
stops before all the units fail. In data with interval censoring, failures occur
between inspection times. For more information on censoring, go to Data censoring.
The type of censoring specifies the information for the table of Test stresses. In the table columns enter one
value per row. Numbers and stored constants are valid.
- No censoring
- The calculations assume the test runs until all the units fail. Every
failure time is an exact observation.
- In the table, specify the test stresses. If Percent allocation of units for each test stress includes user defined allocation, then also enter the percent of
units at each stress level.
- Right censoring time
- The calculations assume the test runs for a preset amount of time. The
amount of time can differ for each stress level.
- In the table, specify the test stresses. For each test stress, specify
the time when the test ends for that row in Censor time. If Percent allocation of units for each test stress includes user defined allocation, then also enter the percent of
units at each stress level.
- Right censoring percent of failures
- The calculations assume the test runs until a preset proportion of
failures occur. The proportion of failures can differ for each stress
level.
- In the table, specify the test stresses. For each test stress, specify
the percent of failures when the test ends for that row in Percent of failures. If Percent allocation of units for each test stress includes user defined allocation, then also enter the percent of
units at each stress level.
- Interval censoring
- The calculations assume that inspections for failures occur at preset
intervals. Each failure falls within an interval of time. You can space
these intervals equally in time or in log time, or you can set intervals
so that the expected number of failures in each interval is the same.
- Equally spaced
- Select to equally space the inspection times. This option is
usually convenient from an operational standpoint because
the inspection schedule is easy to remember.
- For each test stress, enter the number of inspections and
the final inspection time. If Percent allocation of units for each test stress includes user defined allocation, then also enter the
percent of units at each stress level.
- Equal probability
- Select to make the expected proportion of failures the same
in each interval. For a distribution with an increasing
hazard rate, the inspections will be closer together as the
test proceeds. For a distribution with a decreasing hazard
rate, the inspections will be farther apart as the test
proceeds. Usually, you select an option that provides more
inspections when failures are more common.
- For each test stress, enter the number of inspections and
the total percent of failures. If Percent allocation of units for each test stress includes user defined allocation, then also enter the
percent of units at each stress level.
- The total percent of failures is the percent of failures to
observe for the group before the test ends. For example, to
plan that the test for the group ends after 50% of the units
fail, enter 50.
- Equally spaced in natural log times
- Select to equally space the inspections on a natural log
scale. Equal space on the natural log scale produces
inspections that are closer together at the beginning of the
test, then become more spread out as the test proceeds. This
characteristic is usually for failures from a distribution
with a decreasing hazard function. The lognormal
distribution has a decreasing hazard function. A Weibull
distribution with a shape parameter less than 1 has a
decreasing hazard function. Inspections that are closer
together early give increased resolution for early failures.
The less frequent later inspections capture less common
later failures without requiring inspections as
frequently.
- For each test stress, enter the number of inspections and
the first and final inspection times. If Percent allocation of units for each test stress includes user defined allocation, then also enter the
percent of units at each stress level.
