More than 25 years of extreme value statistics for defects: Fundamentals, historical developments, recent applications

A quarter of century has passed since the seminal paper ’94 by Murakami, who launched the adoption of extreme value statistics for the estimation of critical defect size to be considered in fatigue design. Since those early times the discipline, to which the author contributed and witnessed some of the developments, has improved and it has deployed its full potential for the analysis of fatigue properties of AM materials and components. This paper initially deals with the fundamentals of the concepts of extreme value statistics and it presents their historical evolution, driven by real technical problems, to the Peak Over Threshold approach, analysis of rare defects and multiple defect types. The paper then concentrates on applications to AM materials and the open points that still need research and contributions.