ReliaSoft.com > Reliability Software > Weibull++ > Examples > Example 8

Weibull++ Example 8 - Non-Parametric Recurrence Data Analysis

Background

Repairs are tracked on automatic transmissions in a sample of 34 cars in a preproduction road test. The objective is to determine the mean cumulative number of repairs per car by 24,000 test miles (equivalently 5.5 x 24,000 = 132,000 customer miles).

Experiment and Data

(The data for this example appears on page 4 in Recurrent Events Data Analysis for Products Repairs, Disease Recurrences, and Other Applications, by Wayne B. Nelson, 2003.)

The data set shown in the following table consists of each car’s ID, its test mileage at each transmission repair (F) and its latest observed mileage (E). The test was accelerated by a factor of 5.5.

Car Mileage (+ latest)
24 7068   26744 +    
26 28   13809 +    
27 48   1440   29834 +
29 530   25660 +    
31 21762 +        
32 14235 +        
34 1388   21133 +    
35 21401 +        
98 21876 +        
107 5094          
108 21691 +        
109 20890 +        
110 22486 +        
111 19321 +        
112 21585 +        
113 18676 +        
114 23520 +        
115 17955 +        
116 19507 +        
117 24177 +        
118 22854 +        
119 17844 +        
120 22637 +        
121 375   19607 +    
122 19403 +        
123 20997 +        
124 19175 +        
125 20425 +        
126 22149 +        
129 21144 +        
130 21237 +        
131 14281 +        
132 8250   21974 +    
133 19250   21888 +    

Analysis

Step 1: Add a Non-Parametric RDA Folio to an existing project by selecting Add Specialized Folio then Add Non-Parametric RDA from the Project menu.

Step 2: Enter the data into the Data Sheet, as shown next.

Non-Parametric RDA Folio Data Sheet

Step 3: Click Calculate to calculate the Mean Cumulative Function of the data set, then click Show Analysis Summary to display the results in the Results Panel, as shown next.

Results Panel

Step 4: Plot the mean cumulative function on a Plot Sheet, as shown next.

Mean Cumulative Function Plot

From the plot, it can be estimated that the mean cumulative number of repairs per car by 24,000 test miles is 0.3.