A transport framework for zero-variance Monte Carlo estimation of Markovian unreliability

J. Devooght, Pierre-Etienne Labeau, Jean-Luc Delcoux

    Research outputpeer-review


    Monte Carlo simulation has become an important tool for the estimation of reliability characteristics. However it may lead to unacceptable computation times. Variance reduction techniques have to be worked out. We show in this paper how to write the equations of Markovian reliability as a transport problem, and how the well-known-zero-variance scheme can be adapted to this application. But such a method is always specific to the estimation of one quantity, while a Monte Carlo simulation allows to perform simultaneously estimations of diverse quantities. We propound a method to reduce simultaneously the variance for several quantities, by using probability laws that would lead to zero-variance in the estimation of a mean of these quantities. Just like the zero-variance one, the method we propound is impossible to perform exactly. However we show that the use of simple approximations of it may be very efficient.
    Original languageEnglish
    Title of host publicationProbabilistic safety assessment and management
    Place of PublicationLondon
    PublisherSpringer Link
    Number of pages7
    ISBN (Electronic)978-1-4471-3409-1
    StatePublished - 1996
    Event1996 - ESREL PSAM III : Probabilistic Safety Assessment and Management ’96 - Crete
    Duration: 24 Jun 199628 Jun 1996


    Conference1996 - ESREL PSAM III
    Abbreviated titleESREL '96

    Cite this