• Prior knowledge represented in the form of a prior uncertainty distribution
  
  
• Run-time and demand-based data as well as expert judgments

By applying Bayes Theorem, this information is used to compute an updated (posterior) uncertainty distribution reflecting the combined information of the prior distribution, data, and expert judgment. Posterior distributions provide the user not only with a best estimate of the failure rate or failure probability, but also indicate the uncertainty associated with these estimates.

Prior distributions models include uniform, loguniform, normal, lognormal, Beta and Gamma distributions, as well as tabular distributions (probability histograms). This includes generic distributions resulting from non-homogeneous analyses.

Likelihood functions include Binomial, Poisson, for the analysis of demand-based and time-based evidence respectively, as well as additive and multiplicative error models for the incorporation of expert judgments into your analysis.

R-DAT's fast and powerful numerical integration routines algorithms allow almost any combination of prior distribution and likelihood functions, and even combinations of data and expert judgments, to be analyzed. This includes such popular combinations as lognormal prior distributions and Poisson or Binomial likelihood functions.

Prior and posterior distributions are shown both graphically and in tabular form. Both charts and tables can be copied to the Windows clipboard, which allows them to be transferred to other programs such as Microsoft Word and Microsoft Excel.