Weibull++ Example 1 - Complete and Right Censored Data Analysis
Background
Ten identical units are reliability tested at the same application and operation stress levels for 120 hr. The objective is to use the complete and right censored data from the test to determine the unreliability for a mission duration of 226 hr and the warranty time for a reliability of 85%.
Experiment and Data
Six of the units fail during the test after operating for the following numbers of hours: 16, 34, 53, 75, 93 and 120. Four other units are still operating, i.e. right censored or suspended, after 120 hr.
Analysis
Step 1: Using Weibull++ 7, the first step is to create a new life data analysis folio for non-grouped times-to-failure with suspensions data, as shown next.
Step 2: Once the project is created, the next step is to enter the data in the standard folio. The Data Entry Spreadsheet contains a column (State F or S) to indicate whether each data point represents a failure or suspension. When entering data, the user can type an F or S into this column for each data point or use the shortcut provided by Weibull++. Positive values entered into the Time F or S column are automatically marked as failures (F) and negative values are marked as suspensions (S).
The data set is then analyzed using the 2-parameter Weibull distribution and rank regression on X. The results are displayed next.
Step 3: Once the parameters for the data are calculated, several methods were used to find the solution to this case. The first, and more laborious, method is to extract the information directly from a probability plot within Weibull++.
Using RS Draw to mark the coordinates of the intersection of the probability plot line with the expected mission duration and reading the unreliability from the y axis of the plot, the unreliability of the product is determined to be 82%. RS Draw's position indicator shows the unreliability on the probability plot, as shown next.
Step 4: The second method used to determine the unreliability (probability of failure) involves using the Quick Calculation Pad (QCP).
The following figure shows the probability of failure calculated with the QCP, which agrees with the result found using the probability plot.
The next figure shows how the QCP can also be used to determine a warranty time of 32.1482 hr for a reliability of 85%.
