Evaluation of product failure rates for litigation
1Biomedical Engineering Department, Purdue University, West Lafayette, IN 47907-1296, USA.
Abstract:
Two classes of products can be purchased: 1) a product that has been tested and 2) one that has not been tested, but randomly selected samples of the production lot have been subjected to a performance test more demanding than that of the advertised performance level. Based on the results of such random testing, the decision is made to release units in the lot, including the untested units, for sale. An important aspect of this test procedure is careful examination of the range of performance levels of the tested samples, represented by the distribution curve of the numbers of units having specific performance levels versus the performance level. Even with stringent test requirements for the randomly tested samples, there will be a small number of product failures. Hence, an important question is how to draw engineering, scientific, and legal conclusions about the failure of a product that was not itself tested but was part of a batch of products from which a random sample was tested. A methodologic approach for dealing with this situation is presented.
Related Concept Videos
Determination of Expected Frequency
Clinical Trials
There are four phases in a clinical trial. A phase one...
Fatigue
Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast, controlled...
Drug Dissolution: Requirements and Profile Comparison
Clinically Relevant Drug Product Specifications: Methods of Establishment


