Analysis of Accelerated Life Test Data - Part I: The Arrhenius Model and Graphical Methods

This is Part I of a three-part series presenting statistical methods for planning and analyzing temperature-accelerated life tests when all test units are run to failure. These methods are presented so they can be profitably used by individuals with a limited statistical background. In Part I, the Arrhenius model is described, and graphical methods for analysis of such complete data are given. In Part II, numerical methods for analysis of such data are given, and optimum and standard test plans are presented and compared. In Part III, graphical and numerical methods for comparing different products are given, and methods for assessing the validity of the data and the assumptions of the Arrhenius model are also given. These methods are illustrated throughout with accelerated life test data on insulation. While the methods are presented here with the Arrhenius model, they can be used for planning and analyzing many other accelerated life test situations.