Reliability Testing – by Alexander Wassell 11-30-13
Sudden cardiac arrest is a health problem, claiming over 450,000 lives every year in the US. Sudden cardiac arrest is reversible, but only if treated within minutes with the aid of an electrical cardioverter shock via an automated external defibrillator (AED) or with an AED defibrillator. (Sudden Cardian Arrest, n.d.). Sudden Cardiac Arrest (SCA) is an electrical problem caused by electrical malfunction of the heart that results in no blood flow to the body and brain usually caused by an abnormal heart rhythm called ventricular fibrillation (VF).The most effective way to treat Sudden Cardiac Arrest (SCA) is through defibrillation. Medtronic has been able to provide defibrillation through the invention of cardiac defibrillators. Their product is the Medtronic Physio Control LIFEPAK 15 Defibrillator/Monitor, delivers an electrical shock to the heart in order to restore a normal heartbeat within 10 minutes in order to survive an SCA event. If defibrillation does not occur within the window of 10 minutes, the rate of survival drops to less than 5%. LIFEPAK defibrillator's estimated lifecycle is between 3 to 7 years, depending on a number of variables. This wide range prompted Medtronic’s Cardiac Rhythm Disease Management (CRDM) department to request from the engineering department a life data analysis for the product line (times-to-failure). The request, “What is your best, relatively quick, estimate of the reliability of the system over a period of time relevant to the system?”
The desire to obtain reliability results rapidly than when the data comes from products operating under normal conditions via modeling the periods of the product life is a complicated problem. After socialization with the engineering team, the best engineering judgment consensus was to select Accelerated Life Testing (ALT) as the method to quantify the life characteristics and make predictions about all of the units in the population of the LIFEPAK product. Other advantages were reduced test times, reduced time to market, lower product development costs and lower warranty costs. The engineering team conducted standard life data analysis to estimate the parameters for the Weibull life distribution that fit the data at each stress level. The team selected accelerated life test stresses to accelerate the failure modes under consideration and not introduce failure modes that would never occur under use conditions. Stress levels chosen were inside the design limits but outside the product specification limits.
The reliability of the system over a seven years period relevant to the lifecycle was determined to be 81.2%. Fitted to the Arrhenius Acceleration test model for the life of the product, it was determined from the table of percentiles in the probability plot that the confidence interval was large and not credible due to the low number of tests conducted. The Mean Time to Failure (MTTF) was determined to...