Here is the abstract you requested from the medical_2018 technical program page. This is the original abstract submitted by the author. Any changes to the technical content of the final manuscript published by IMAPS or the presentation that is given during the event is done by the author, not IMAPS.
|Accelerated testing and predicted useful lifetime of medical electronics|
|Keywords: design for reliability, predictive modeling, remaining lifetime|
|Understanding the physics-of-failure is critical to the assurance of the medical electronics (ME) reliability [1,2], and is imperative for the prediction of the remaining useful lifetime (RUL)  of a ME device. The recently suggested probability design for reliability (PDfR) concept [4,5], with its experimental basis in the highly-focused and highly-cost- effective failure oriented accelerated testing (FOAT) [6-8] and effective and flexible Boltzmann-Arrhenius-Zhurkov (BAZ) predictive model [9-14], can be effectively used for the prediction, on the probabilistic basis , the RUL of the electronic product of interest. It is shown also 1) how the physics-of-failure-based BAZ model can be sandwiched between two statistical models, when there is a need to diagnose a presumably faulty device (Bayes theorem can be used to do that), assess its RUL (using BAZ model) and update the product’s reliability (using beta-distribution) , and 2) how to assess, for a medical device composed by mass- produced components, the favorable effect of the statistics-of-failure related (SFR) process (that results in the decreasing failure rate with time) interacting with the physics-of- failure related (PFR) degradation (aging) process (that results in the increasing failure rate with time) . The general concepts are illustrated by practical numerical examples.|
|Ephraim Suhir, Research Professor
Portland State University, Portland, OR, USA; Technical University, Vienna, Austria; and ERS Co., Los Altos, CA, 94024, USA,
Los Altos, CA