Partially accelerated life testing models based on censored samples / Hamdy Goma Ahmed Aboelelaa ; Supervised Amal Soliman Hassan , Marwa Abdalla Abdelghafar

Thesis (M.Sc.) - Cairo University - Institute of Statistical Studies and Research - Department of Mathematical Statistics

The continuous improvement in manufacturing design creates a problem in obtaining information about lifetime of some products and materials with high reliability at the time of testing under normal conditions. Under such conditions the life testing becomes very expensive and time consuming. To obtain failures quickly, a sample of these materials is tested at more severe operating conditions than normal ones. In these cases, accelerated tests can be applied to reduce the experimental time and hence the cost. Accelerated life testing (ALT) is a quick way to obtain information about the life distribution of a material, component or product. The fundamental assumption in ALT is that the mathematical model relating the lifetime of the unit and the stress should be known or can be assumed. In some cases, this kind of life-stress relationships are not known and also cannot be assumed, i.e. ALT data cannot be extrapolated to use condition. So, in such cases, partially accelerated life tests (PALT) is a more suitable test to be performed for which tested units are subjected to both normal and accelerated conditions. This thesis concerns with estimation problem in step{u2013}stress partially accelerated life test (SSPALT) based on Type II progressive censoring with random removal. The parameter estimation is considered for two important life time models (exponentiated inverted Weibull and exponntiated Pareto). The point estimates and approximate confidence intervals for the parameters of both models are obtained. Numerical study is performed to assess the theoretical results

Issued also as CD