Enter your criteria for Accelerated Life Test Plan

Stat > Reliability/Survival > Test Plans > Accelerated Life Testing

Enter planning information to determine the number of test units and how to allocate the units across stress levels for an accelerated life test.

Note

The data you collect for an accelerated life test can be censored or uncensored. A time-censored test plan or failure-censored test plan often provides precise results while minimizing test costs. For more information, go to Data censoring.

In This Topic

Calculate
Relationship between life and stress
Specify test stresses
Distribution assumptions
Specify planning values for regression coefficients
Sample sizes or Precisions as distances

Calculate

Select whether you want the results to provide sample sizes or to provide precision values in relation to a confidence bound.

Sample sizes: If you have a precision goal, select Sample sizes to determine the plan that requires the minimum number of samples needed to obtain that precision.
Standard errors of estimates: If you have a certain number of units available for testing, select Standard errors of estimates to determine the plan that provides the best precision for the sample size.

Relationship between life and stress

Select whether to transform the accelerating variable. If you can assume a linear relationship, select Linear and no transformation is performed. To transform the accelerating variable, select Arrhenius, Inverse temp, or Ln (Power).

Note

For information on the distribution models and the transformations for the accelerating variable, go to Choose the appropriate model for accelerated life testing.

Specify test stresses

Select Specify test stresses to enter the design stress and the stresses for the test. The optional information to specify includes censoring and allocations. For more information, go to Specify stress levels for Accelerated Life Test Plan.

Distribution assumptions

Under Distribution assumptions, provide information about a distribution to model your data.

Base your decision on process knowledge or an evaluation of the distribution fit. For more information, go to Distribution fit for reliability analysis.

The planning values depend on the distribution.

Weibull

Planning shape parameter value: Enter a positive number. For more information on the Weibull distribution and its parameters, go to Weibull distribution in reliability analysis.
Assume shape parameter is known: If you select this option, Minitab assumes that you do not need to estimate this parameter. Therefore, a smaller sample size is required.

Exponential

No distribution assumptions are necessary for the exponential distribution. The test plan assumes a constant hazard function.

Other distributions

Planning scale parameter value: Enter a positive number. The scale parameter relates most strongly to the variation in the data. Larger values indicate more variation. The relationship between the scale parameter and the data depends on the distribution.

Specify planning values for regression coefficients

Select which values to provide. The values specify the line for the relationship between the accelerating variable and time. For example, if you specify percentile points for two distinct stress levels, the analysis forms a line that connects those two points.

If you have previously used accelerated life tests for similar experiments, you can use historical estimates of the slope and intercept as planning values. The interpretation of the slope and intercept depends on the relationship between life and stress. For more information on the slope and intercept of accelerated test models, go to Choose the appropriate model for accelerated life testing.

Consider the following information to enter each type of value.

Percentile point

Enter three values to define a percentile point.

Percentile value: Enter the time value when the percent of failures occurs for the percent. The value assumes the stress level that you specify.
Percent: Enter a value between 0 and 100. The percent is the percent of failures at the percentile value. The value assumes the stress level that you specify.
Stress level: Enter the value of the accelerating variable that produces the percentile value and percent.
Note
If you specify planning values for two percentiles, they must be at different stress levels. For more information, go to Values for estimation test plans and accelerated life test plans.

For example, if you expect 50% of electrical components to fail after 450 hours at a voltage of 7V, enter 450 in Percentile, 50 in Percent, and 7 in Stress.

Intercept value: Enter the intercept for the relationship with the accelerating variable. The intercept is the value of the relationship when the x value is 0 after the transformation for the accelerating relationship.
Slope value: Enter the slope for the relationship with the accelerating variable.; The slope represents the activation energy when the Arrhenius relationship is chosen and the assumed distribution is Weibull, exponential, lognormal, or loglogistic.

Sample sizes or Precisions as distances

Enter sample sizes or precision values. The value that you enter depends on the value to calculate.

Sample sizes

Enter one or more values to indicate the number of units available for testing.

Precisions as distances

Select Lower confidence bound to estimate or Upper confidence bound to estimate. The lower confidence bound is commonly used because many reliability standards are defined in terms of the worst-case scenario. For example, if you want to estimate the 10th percentile of your failure time distribution, and the lower confidence bound is to be no more than 25 hours less than your estimate, then select Lower bound and enter 25 in Precision values.

Precision values: Enter one or more positive values. The values are the distance from the reliability estimate to the bound of the confidence interval (CI). The analysis produces separate plans for each precision.