A0174
Title: Extended gamma process model for accelerated destructive degradation analysis
Authors: Man Ho Ling - The Education University of Hong Kong (Hong Kong) [presenting]
Abstract: Accelerated degradation destructive testing (ADDT) has emerged as a valuable technique for analyzing highly reliable products. This approach has garnered significant attention in the field of reliability research. A common characteristic observed in many degradation studies is the presence of randomness in the initial degradation levels of tested units. Products with lower initial degradation levels tend to experience failure at an earlier stage. In light of this, an extended gamma process model specifically designed to analyze ADDT data with random initial degradation levels is presented. Approximations for the conditional mean-time-to-failure and the variance of failure time for products exhibiting higher degradation levels are also presented. These approximations provide valuable insights to practitioners in evaluating the impacts of different degradation levels on product reliability. Finally, an ADDT dataset of return springs illustrates the proposed model and methodologies for making informed decisions regarding quality management and product performance.
