Product Component Genealogy Modeling and Warranty Return Prediction

• Activity Number: 243
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Section on Physical and Engineering Sciences
• Author(s): Caleb B. King*+ and Yili Hong and William Q. Meeker
• Companies: Virginia Polytechnic Institute & State University and Virginia Polytechnic Institute & State University and Iowa State University
• Keywords: Competing risks ; Component generation ; Lifetime data ; Multiple failure modes ; Prediction intervals ; Product reliability

Abstract

Many industrial products consist of multiple components that are necessary for maintaining their integrity. There is already an abundance of literature on modeling the lifetime of such components through competing risks models. However, none of these models take into consideration the fact that many of these components are part of a specific generation of components and that each of these components may go through several generations throughout a period of observation. We believe that by incorporating this extra information, which is readily available in most product warranty databases, better accuracy can be achieved in predicting time to failure thus yielding more reliable warranty return information. We present methods for estimating parameters for this new generational model as well as provide a comparison with existing methods through the use of simulation. Our results indicate that the generational model outperforms the existing methods in predicting warranty returns.