Interpretation summary for Nonparametric Distribution Analysis (Arbitrary Censoring)

Use nonparametric distribution analyses to provide estimates that describe the reliability of your product for the following conditions:

When no known distribution adequately fits your data.
When you want to validate or compare the results from a parametric distribution analysis.

Note

Parametric methods provide more precise results and estimate more types of functions.

Depending on which nonparametric method you choose, you can do the following:

Display distribution characteristics such as median time to failure
Estimate percentiles and survival probabilities
Display survival and hazard plots

When your product fails in different ways, use a failure mode analysis to evaluate the impact of each type of failure on the overall reliability. Each failure mode is assumed to be independent. By analyzing each failure mode individually, you can more easily prioritize your improvement efforts.

You can use a failure mode analysis to assess the following:

Reliability of each failure mode
Overall product reliability
Overall product reliability after removing a specific failure mode

Data description

Muffler Data: Single failure mode analysis

MufflerReliability.MTW

A company set the warranty of its new type of mufflers at 50,000 miles. The reliability group wants to assess the reliability of the new type of mufflers and to estimate the proportion of warranty claims they can expect. The measurement of interest is the number of miles driven on a muffler when a warranty claim is made.

The reliability group performs a nonparametric distribution analysis using both the Turnbull and actuarial methods.

Pump Data: Multiple failure mode analysis

WaterPumpReliability.MTW

The same automobile manufacturer is also working with the supplier of their water pumps to improve their overall reliability. To determine how to focus the improvement effort, engineers collect data on how and when water pumps fail.

The engineers perform a nonparametric multiple failure mode analysis using the Turnbull method.