Home / Statistical Tools / Reliability / Censored / Reliability: Censored vs. Uncensored Data

Reliability: Censored vs. Uncensored Data

Quantum XL can model both uncensored and censored data. If you've tested all your units to failure, then use uncensored fitting. If some of the units were removed from testing prior to failure, then use censored fitting. For more information about censored modeling, see Understanding Censored Data.

Uncensored Example

100 units were put on life test. The time to fail for all 100 units is recorded and ready for distribution fitting. The time to fail for the first five units are below. Note: since there were 100 units put on test, the dataset should include 100 numbers.

Uncensored data can be fit using either Auto-Fit or choosing a distribution manually.

Censored Example

52 units were put on life test and inspected approximately every five days. If a unit was found defective during an inspection, the exact time of failure was not known, only that it failed sometime between inspections. Additionally, at the end of 20 days, 45 units had not failed and were removed from life test. Below is an example dataset for censored reliability data.

For more information about censored modeling, see Understanding Censored Data.

Censored data can be fit using either Auto-Fit or choosing a distribution manually.