Chapter Contents Previous Next
 The RELIABILITY Procedure

## Weibull Analysis of Interval Data with Common Inspection Schedule

Table 30.1 shows data for 167 identical turbine parts provided by Nelson (1982, p. 415). The parts were inspected at certain times to determine which parts had cracked since the last inspection. The times at which parts develop cracks are to be fitted with a Weibull distribution.

Table 30.1: Turbine Part Cracking Data
 Inspection (Months) Number Start End Cracked Cumulative 0 6.12 5 5 6.12 19.92 16 21 19.92 29.64 12 33 29.64 35.40 18 51 35.40 39.72 18 69 39.72 45.24 2 71 45.24 52.32 6 77 52.32 63.48 17 94 63.48 Survived 73 167

Table 30.1 shows the time in months of each inspection period and the number of cracked parts found in each period. These data are said to be interval censored since only the time interval in which failures occurred is known, not the exact failure times. Seventy-three parts had not cracked at the last inspection, which took place at 63.48 months. These 73 lifetimes are right censored, since the lifetimes are known only to be greater than 63.48 months.

The interval data in this example is read from a SAS data set with a special structure. All units must have a common inspection schedule. This type of interval data is called readout data. The SAS data set named CRACKS, shown in Figure 30.8, provides the data in Table 30.1 with this structure. The variable TIME is the inspection time, that is, the upper endpoint of each interval. The variable UNITS is the number of unfailed units at the beginning of each interval, and the variable FAIL is the number of units with cracks at the inspection time.

 Obs time units fail 1 6.12 167 5 2 19.92 162 16 3 29.64 146 12 4 35.40 134 18 5 39.72 116 18 6 45.24 98 2 7 52.32 96 6 8 63.48 90 17
Figure 30.8: Listing of the Turbine Part Cracking Data

The following statements use the RELIABILITY procedure to produce the probability plot in Figure 30.9 for the data in the data set CRACKS.

```   symbol v=plus c=blue;
proc reliability data=cracks;
freq fail;
nenter units;
distribution weibull;
pconfplt
noconf
cframe = ligr;
inset / ctext = black
cfill = ywh ;
run;
```
The FREQ statement specifies that the variable FAIL provides the number of failures in each interval. The NENTER statement specifies that the variable UNITS provides the number of unfailed units at the beginning of each interval. The DISTRIBUTION statement specifies that the Weibull distribution is used for parameter estimation and probability plotting. The PROBPLOT statement requests a probability plot of the data.

The PROBPLOT statement option READOUT indicates that the data in the CRACKS data set are readout (or interval) data. The option PCONFPLT specifies that confidence intervals for the cumulative probability of failure are to be plotted. The confidence intervals for the cumulative probability are based on the binomial distribution for time intervals until right censoring occurs. For time intervals after right censoring occurs, the binomial distribution is not valid, and a normal approximation is used to compute confidence intervals.

The option NOCONF suppresses the display of confidence intervals for distribution percentiles in the probability plot.

Figure 30.9: Weibull Probability Plot for the Part Cracking Data

A listing of the tabular output produced by the preceding SAS statements is shown in Figure 30.10. By default, the specified Weibull distribution is fitted by maximum likelihood. The line plotted on the probability plot and the tabular output summarize this fit. For interval data, the estimated cumulative probabilities and associated confidence intervals are tabulated. In addition, general fit information, parameter estimates, percentile estimates, standard errors, and confidence intervals are tabulated.

 The RELIABILITY Procedure

 Model Information Input Data Set WORK.CRACKS Analysis Variable time Frequency Variable fail NENTER Variable units Distribution Weibull Estimation Method Maximum Likelihood Confidence Coefficient 95% Observations Used 8

 Cumulative Probability Estimates Lower Lifetime Upper Lifetime CumulativeProbability 95% Confidence Limits Standard Error Lower Upper . 6.12 0.0299 0.0098 0.0685 0.0132 6.12 19.92 0.1257 0.0796 0.1858 0.0257 19.92 29.64 0.1976 0.1401 0.2662 0.0308 29.64 35.4 0.3054 0.2366 0.3813 0.0356 35.4 39.72 0.4132 0.3376 0.4918 0.0381 39.72 45.24 0.4251 0.3491 0.5039 0.0383 45.24 52.32 0.4611 0.3838 0.5398 0.0386 52.32 63.48 0.5629 0.4841 0.6394 0.0384

 Algorithm converged.

 Summary of Fit Observations Used 8 Right Censored Values 73 Left Censored Values 5 Interval Censored Values 89 Maximum Loglikelihood -309.6684

 Weibull Parameter Estimates Parameter Estimate Standard Error Asymptotic Normal 95% Confidence Limits Lower Upper EV Location 4.2724 0.0744 4.1265 4.4182 EV Scale 0.6732 0.0664 0.5549 0.8168 Weibull Scale 71.6904 5.3335 61.9634 82.9444 Weibull Shape 1.4854 0.1465 1.2242 1.8022

 Other Weibull DistributionParameters Parameter Value Mean 64.7966 Mode 33.7622 Median 56.0144

 Weibull Percentile Estimates Percent Estimate Standard Error Asymptotic Normal 95% Confidence Limits Lower Upper 0.1 0.68534385 0.29999861 0.29060848 1.61625083 0.2 1.09324674 0.42889777 0.50673224 2.3586193 0.5 2.02798319 0.67429625 1.05692279 3.8912169 1 3.23938972 0.93123832 1.84401909 5.69063837 2 5.18330703 1.2581604 3.22101028 8.34106988 5 9.70579945 1.78869256 6.76335893 13.9283666 10 15.7577991 2.22445157 11.9491109 20.7804776 20 26.1159906 2.6327383 21.4337103 31.821134 30 35.8126238 2.90557264 30.547517 41.9852137 40 45.6100472 3.27409792 39.6239146 52.5005271 50 56.0143651 3.89410377 48.8792027 64.1910859 60 67.5928125 4.90210777 58.6364803 77.917165 70 81.2334227 6.46932648 69.4938134 94.9562075 80 98.7644937 8.95137184 82.6900902 117.963654 90 125.694556 13.5078386 101.821995 155.164133 95 150.057755 18.2060035 118.300075 190.340791 99 200.437864 29.1957544 150.658574 266.66479 99.9 263.348102 44.7205513 188.791789 367.347666
Figure 30.10: Listing of the Tabular Output for the Part Cracking Data

In this example, the number of unfailed units at the beginning of an interval minus the number failing in the interval is equal to the number of unfailed units entering the next interval. This is not always the case since some unfailed units might be removed from the test at the end of an interval; that is, they might be right censored. The special structure of the input SAS data set required for interval data enables the RELIABILITY procedure to analyze this more general case.

 Chapter Contents Previous Next Top