Turnbull estimation method for Nonparametric Distribution Analysis (Arbitrary Censoring)

Probability of failure – Turnbull estimation method

The probability of failure provides, for each interval, the chance that the product will fail in that interval. Use this information to determine the following:
  • Which intervals have the most failures
  • Whether the failures are spread among many time intervals or concentrated among a few intervals

Nonparametric estimates do not depend on any particular distribution and therefore are good to use when no distribution adequately fits the data.

Example output

Turnbull Estimates

IntervalProbability
of Failure
Standard
Error
LowerUpper
20000300000.0028600.0016488
30000400000.0104860.0031451
40000500000.0324120.0054678
50000600000.1029550.0093830
60000700000.1706390.0116151
70000800000.2488080.0133481
80000900000.2316490.0130259
90000*0.200191*

Interpretation

For the new muffler data, 0.248808 (or 24.8808%) of the new type of mufflers failed in the interval from 70,000 to 80,000 miles.

Survival probabilities – Turnbull estimation method

The survival probabilities indicate the probability the product survives until a particular time. Use these values to determine whether your product meets reliability requirements or to compare the reliability of two or more designs of a product.

Example output

Table of Survival Probabilities


Survival
Probability
Standard
Error
95.0% Normal CI
TimeLowerUpper
300000.9971400.00164880.9939091.00000
400000.9866540.00354300.9797100.99360
500000.9542420.00645170.9415970.96689
600000.8512870.01098560.8297560.87282
700000.6806480.01439490.6524350.70886
800000.4318400.01529360.4018650.46181
900000.2001910.01235460.1759760.22441

Interpretation

For the new muffler data, 0.954242 (or 95.4242%) of the new type of mufflers survive at least 50,000 miles.