<<< Back to Index
This Month's FMEA Corner >>>

Is there a way to obtain plots and results from a reliability model without entering an actual data set?



Watch this video and others like it on ReliaSoft.tv

Yes, the Weibull++, ALTA and RGA desktop applications all give you the flexibility to generate plots and calculate metrics based on a specified model without entering data.

Note that if you don’t know the model parameters, you can use the Quick Parameter Estimator (QPE) to estimate them based on what you know about the behavior over time.

Option 1: Entering the Parameters in a Blank Data Sheet

First, create a blank standard folio data sheet. On the control panel, choose a distribution/model for the analysis and then click the Calculate icon. Enter the parameters in the input window and click OK.

The control panel of the data sheet will display the parameters of the specified model, and you will be able to use most of the data folio tools (including the QCP and Plot) in the same way that you would for an analyzed data set. Some features, including confidence bounds, will not be available when the data sheet is empty.

Note that in RGA, this option is not available for the Crow Extended and Crow Extended – Continuous Evaluation models, or for data sheets that support the Advanced Systems View.

Using an ALTA standard folio as an example, the following pictures show the Unreliability vs. Time plot and a QPC result for an Arrhenius-Weibull model with parameters Beta = 1.5, B = 5000 and C = 0.001.

Unreliability vs. Time plot and QCP results

Option 2: Generating a Data Set

To generate a data set using simulation, click the applicable Monte Carlo command on the Home tab on the ribbon.

Weibull++ Monte Carlo

Weibull++ Monte Carlo

ALTA Monte Carlo

ALTA Monte Carlo

RGA Monte Carlo

RGA Monte Carlo

In the Monte Carlo Generation window, enter the parameters, number of data points and other information required to define the data set you wish to create, and then click Generate. The data will be generated in a standard folio data sheet, where you can use the data folio tools in the same way that you would for an actual data set. In this case, confidence bounds are available.

The following example shows the data set generated by the Weibull++ Monte Carlo utility for a 2-parameter Weibull distribution with parameters beta = 1.5 and eta = 100, and set at 20 data points with no censoring.

Weibull++ Monte Carlo

Alternatively, Weibull++ offers a second approach for generating a data set using numerical methods. To use this approach, create a blank standard folio, choose a distribution from the control panel and click Calculate.

In the Parameters tab of the input window, enter the parameters of the model. In the Options tab, select the Add data and calculate check box, and then enter the desired sample size, number of failures and the probability value.

Weibull++ Standard Folio Input Window

When you click OK, the software will solve for the unreliability value for each failure, and then use these values to obtain the failure times based on the specified distribution and parameters. You can then use the data folio tools in the same way that you would for an actual data set.

Weibull++ Standard Folio Data Set