Optimum allocations test plans for Accelerated Life Test Plan

Minitab evaluates the resulting test plans and displays the best plans with respect to minimizing the standard error of the parameter of interest. By default, Minitab displays three test plans. The first best optimum allocations test plan is the one with the smallest variance.

Minitab also displays the second and third best optimum allocations of test units. If little difference exists in the standard error of the parameter of interest across test plans, choose the plan that would yield more failures or is the least expensive to execute.

Example output

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

Interpretation

For the capacitor data, the allocation of 100 units that minimizes the variance of the 1000-hour reliability is:

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.

The estimated standard error of the 1000-hour reliability is 0.0123437.