Statistical modeling of creep rupture data

A flexible statistical modeling framework for the analysis elf creep ruptur e data is proposed, which offers an improvement on traditional methods of d eriving creep rupture strength values and confidence limits. The paper revi ews a family of models that can be used to represent the trend relationship between failure time and stress at any temperature, together with the rand om scatter of failure times about the trend line, and examines the reliabil ity of extrapolations. Areas of statistical research which would lead to mo del improvement are discussed, such as variance heterogeneity, left censori ng and allowance for the cluster (cast) structure of the data.