Skip to Main content Skip to Navigation

Modélisation stochastique du vieillissement en temps discret

Abstract : In this thesis, we propose a complete review of non repairable systems reliability concepts and models when systems lifetime are discrete random variables. First, we give the basic concepts of reliability in discrete time, including ageing notions. We show that the differences between continuous and discrete reliability ageing concepts are due to the definition of discrete failure rate. We propose a new definition of failure rate in discrete time which solves the problems. A bibliographical study of discrete reliability models highlights a possible classification of them into three categories. A complete statistical analysis is presented, from a non parametric study of reliability functions involved in the characterization of ageing, to parameters estimation and goodness-of-fit tests for discrete models. We propose a new goodness-of-fit test based on the generalized Smirnov transformation whose power is studied in the case of fitting the geometric and type I discrete Weibull distributions.
Document type :
Complete list of metadatas
Contributor : Thèses Imag <>
Submitted on : Monday, February 16, 2004 - 10:24:20 AM
Last modification on : Friday, November 6, 2020 - 4:13:23 AM
Long-term archiving on: : Friday, April 2, 2010 - 7:41:05 PM


  • HAL Id : tel-00004670, version 1



Cyril Bracquemond. Modélisation stochastique du vieillissement en temps discret. Modélisation et simulation. Institut National Polytechnique de Grenoble - INPG, 2001. Français. ⟨tel-00004670⟩



Record views


Files downloads