SEQUENTIAL LIFE TESTING WITH TRUNCATION MECHANISMS FOR UNDERLYING THREE-PARAMETER WEIBULL AND INVERSE WEIBULL MODELS

摘要

The two-parameter Inverse Weibull distribution was derived by Pascoal Erto [1]. It has a location, a scale and a shape parameter, and it has been used in Bayesian reliability estimation to represent the information available about the scale parameter of an underlying Weibull model (Erto [1]; De Souza & Lamberson [2]). It has also been previously used in sequential life testing situations by De Souza [3], [4]. The location parameter was assumed to be equal to zero. The two-parameter Weibull distribution is widely used as a failure model (De Souza [5], [6], [7]). It has a location, a scale and a shape parameter. The location parameter also was assumed to be equal to zero. It happens that when its shape parameter is greater than 7, the two-parameter Weibull curve becomes highly pointed, resulting in some computational difficulty (accuracy) in calculating the component's characteristics of interest values. In situations like that, the two-parameter Inverse Weibull distribution seems to have a better answer to the accuracy problem presented by the two-parameter Weibull model, as shown by De Souza [4]. In this work we will analyze situations where the underlying sampling distributions were, respectively, the three-parameter Inverse Weibull and three-parameter Weibull models. We will be assuming that the location parameters of both distributions are different from zero. We will also develop truncation mechanisms for both models. An example will Illustrate the application of a sequential life testing approach when the underlying sampling distributions are the three-parameter Inverse Weibull and three-parameter Weibull models.