Reliability Prediction Using Additive Weibull Model

<div class="section abstract"><div class="htmlview paragraph">This paper presents additive Weibull reliability model using customer complaints data and finite element fatigue (FEA) analysis data. Warranty data provides insight into the underlying customer issues. Reliability engineers prepare a prediction model based on this data to forecast the failure rate of components. However, warranty data has certain limitations with respect to prediction modeling. The warranty period covers only the infant mortality and useful life zone of a bathtub curve. Thus, predicting with solely warranty data generally cannot provide results with desired accuracy.</div><div class="htmlview paragraph">The failure rate of wear-out components is driven by random issues initially and wear-out or usage-related issues at the end of the lifetime. For accurate prediction of failure rate, data need to be explored at wear-out zone of a bathtub curve. Higher cost always limits the testing of components until failure, but FEA fatigue analysis can provide the failure rate behavior of a part much beyond the warranty period without physical testing.</div><div class="htmlview paragraph">In this work, the authors proposed an additive Weibull model, which uses both warranty and FEA fatigue life data for predicting failure rates. Prediction model involves two data sets of a part: one with existing warranty claims and other with fatigue life data. Hazard rate base Weibull estimation has been used for modeling the warranty data whereas S-N curved-based Weibull parameter estimation is used for FEA data. To separate Weibull models’ parameters, they are first estimated and combined to form the proposed mix Weibull model.</div></div>