Skip to Main content Skip to Navigation

Modélisation et évaluation de l'efficacité de la maintenance des systèmes réparables

Abstract : This work deals with stochastic modeling and parametric estimation of preventive and corrective maintenance efficiency for complex repairable systems. In the first part, imperfect repair models are classified using the failure intensity of a random point process. This classification includes existing models and proposes new ones. Then, the work focuses on models called arithmetic reduction of age and intensity and on the Brown-Proschan model. New probabilistic results are proposed in order to better understand the behaviour of the failure process of these models and to develop statistical methods for parameters estimation. In the second part, two modeling frameworks are proposed and studied in order to generalize these models to the various types of preventive maintenances existing in the industrial context: systematic, planned or conditional preventive maintenances. The general models proposed allow to include a possible dependency between preventive and corrective maintenance times. Finally, a method for assessing both global preventive and corrective maintenance efficiency is developped and illustrated on a realistic example issued from the industry.
Document type :
Complete list of metadata
Contributor : Laurent Doyen <>
Submitted on : Monday, December 13, 2004 - 11:11:25 AM
Last modification on : Wednesday, July 28, 2021 - 3:57:06 AM
Long-term archiving on: : Thursday, September 13, 2012 - 1:40:29 PM


  • HAL Id : tel-00007722, version 1



Laurent Doyen. Modélisation et évaluation de l'efficacité de la maintenance des systèmes réparables. Mathématiques [math]. Institut National Polytechnique de Grenoble - INPG, 2004. Français. ⟨tel-00007722⟩



Record views


Files downloads