Example of Accelerated Life Test Plan

An engineer tests capacitors at accelerated temperatures of 85, 105, and 125 degrees to determine the 1000-hour reliability under normal-use conditions of 45 degrees. 100 capacitors are available to test. The engineer uses the following information for the test plans.

Failure times tend to follow an exponential distribution.
The Arrhenius relationship with an intercept of -8.0 and slope of 0.5 appropriately models the relationship between log failure time and temperature.

Choose Stat > Reliability/Survival > Test Plans > Accelerated Life Testing.
Under Parameter to be Estimated, select Reliability at time, and enter 1000.
From Sample sizes or precisions as distances from bound of CI to estimate, select Sample size and enter 100.
From Distribution, select Exponential. From Relationship, select Arrhenius.
Under Specify planning values for two of the following, in Intercept, enter -8. In Slope, enter 0.5.
Click Stresses.
In Design stress, enter 45. In Test stresses, enter 85 105 125.
Click OK in each dialog box.

Interpret the results

Minitab evaluates the resulting test plans and displays the best plans with respect to minimizing the precision or the sample size. If the sample size is specified, the 1st Best "Optimum" Allocations Test Plan is the plan with the smallest standard error of the parameter of interest.

To estimate the 1000-hour reliability at the design stress of 45 degrees using the optimal plan, the engineer should test the following number of units at each accelerated temperature:

Test 63 units at 85 degrees
Test 4 units at 105 degrees
Test 33 units at 125 degrees

The settings for this test plan have no censoring for any of the stress levels, so the plan is that the test continues until 100% of the units fail.

Because the standard error for all three plans is very close, the engineer should also consider additional criteria when choosing an allocations plan, such as which plan yields more failures or is the least expensive to execute.

Planning Distribution

Distribution	Intercept	Slope
Exponential	-8	0.5

Uncensored data

Arrhenius model

Estimated parameter: Reliability at time = 1000

Calculated planning estimate = 0.964857

Design stress value = 45

Selected test plans: “Optimum” allocations test plans

Total available sample units = 100

1st Best “Optimum” Allocations Test Plan

Test Stress	Percent Failure	Percent Alloc	Sample Units	Expected Failures
85	100	63.3333	63	63
105	100	3.9583	4	4
125	100	32.7083	33	33

2nd Best “Optimum” Allocations Test Plan

Test Stress	Percent Failure	Percent Alloc	Sample Units	Expected Failures
85	100	68.3333	68	68
105	100	3.9583	4	4
125	100	27.7083	28	28

3rd Best “Optimum” Allocations Test Plan

Test Stress	Percent Failure	Percent Alloc	Sample Units	Expected Failures
85	100	58.3333	58	58
105	100	8.9583	9	9
125	100	32.7083	33	33