A Study of Degradation Data with Measurement Errors

• Activity Number: 612
• Type: Contributed
• Date/Time: Wednesday, August 12, 2015 : 2:00 PM to 3:50 PM
• Sponsor: Quality and Productivity Section
• Author(s): Chien-Yu Peng*
• Companies: Institute of Statistical Science, Academia Sinica
• Keywords: gamma process ; inverse Gaussian process ; random effect ; Wiener process

Abstract

Degradation models are widely used to assess the lifetime information of highly reliable products. When there are measurement errors in monotonic degradation paths, non-monotonic degradation model assumptions may lead to contradictions between physical mechanisms and statistical explanations. This study presents a Levy degradation-based process which simultaneously considers the unit-to-unit variability, within-unit variability and measurement error in the degradation data. In addition, this paper uses a Monte Carlo method to estimate the model parameters and provides a simple model-checking procedure to assess the validity of model assumptions. Finally, an example is performed to illustrate the flexibility of the proposed models.