Journal of Quality Engineering and Management

Abstract The reliability of manufactured products can vary according to changes in production quality. Field failure data provide useful information for assessing whether changes in reliability are significant or identifying the cause of changes. In order to identify these errors, we need to model the effect of these errors on product reliability. In this research, we intend to predict product reliability behavior based on the percentage of different quality errors with which products may be produced. In this regard, two types of quality errors, namely non-compliant items and assembly error are examined separately. In order to model, it is assumed that the percentage of qualitative errors follow the beta distribution and the failure times follow the Weibull distribution. Reliability, risk rate and probability chart of products are studied under these two types of qualitative errors. Based on the results of this research, it is possible to guess the type and percentage of quality errors with which products are produced.

Abstract ehavior is a function of the failure rate of some mechanical systems. The conventional Weibull model is not able to fully model the lifespan of such systems with the ohmic refractive index function. In this paper, a new generalized Weibull distribution is used to model the failure rate functions with vanity-shaped behavior. This model is evaluated on three datasets. In these three datasets, in order to reduce the execution time of the lifetime test, accelerator and censored type one tests have been used. The model parameters are also estimated based on the maximum likelihood method. The Akaike indices, the logarithm of the probability function, and the Bayesian information criterion are obtained, indicating the efficiency of this model for the data obtained from performing the lifetime tests. Therefore, the predicted average lifespan has good validity.