%0 Journal Article
%T The Problem of Sprinkler Reliability Sprinkleri  patikimumo problema
%A Jurgita £żak£żnait£ż
%J Science ¨C Future of Lithuania
%D 2009
%I Vilnius Gediminas Technical University
%R 10.3846/186
%X Sprinkler systems allow a considerable reduction of fire risk in buildings. Unfortunately, sprinklers are not fail-safe technical systems. Relatively high rates of sprinkler failures evoke the problem of reliability. A solution to this problem is considered from several viewpoints. The diversity of sprinklers' failure modes is the first challenge for estimating reliability (failure probability). It is found that the use of the available data for estimation is problematic. The second challenge is that the published data is insufficiently described to allow a verification of its relevance to the specific case of failure probability estimation. It is suggested to apply the published data with partial relevance to Bayesian inference about failure probabilities. The data is used for developing prior distributions of the unknown values of the probabilities. Bayesian inference is carried out on the basis of binomial distribution used to model the operation of sprinklers on demand basis. A problem of aging and a possible increase in failure probability in the course of sprinkler service is shortly discussed.  Article in English Nagrin ta sprinkleri  sistem  patikimumo vertinimo problema. Patikimumo matu laikoma sprinkleri  atsako tikimyb . Rasta, kad  ios tikimyb s vertinimas yra s lygi kai sud tingas u davinys. Sprinkleriai gali patirti atsak  labai  vairiais b¨ħdais. Be to,  is atsakas patiriamas retai. Tai lemia duomen  apie sprinkleri  ne sijungimo gaisro metu negausum , neleid ia vertinti atsako tikimyb s, remiantis klasikiniu santykinio da nio po i¨ħriu. Patikimumo vertinim  komplikuoja tai, kad duomenys apie sprinkleri  atsakus renkami gana nesistemingai, be teorinio pagrindo ir standartizavimo  alies, regiono ar pasaulio mastu. Duomenys paprastai fiksuojami pasyviai, kai sprinkleriai ne sijungia gaisro metu. Turimi empiriniai sprinkleri  atsako da niai gali b¨ħti naudojami kaip  iurk t¨ħs tikimybi   ver iai. Juos galima taikyti Bajeso analizei, kai reikia formuoti apriorin  atsako tikimyb s tank .    tank  galima atnaujinti pasitelkiant negausius duomenis, surinktus konkre ioje  alyje, ir gautus stebint konkretaus tipo sprinklerius. Teorinis modelis kuriuo galima nusakyti sprinkleri  atsak  tik tinum  per pasirinkt  gaisr  skai i , yra binominis skirstinys. Sprinkleri  atsako tikimyb s vertinim  galima atlikti taikant standartin  binominio skirstinio parametro vertinimo proced¨ħr  Bajeso statistikos metodais. Pagrindin  binominio skirstinio prielaida yra ta, kad sprinkleriai nesensta ir j  atsako tikimyb  nesikei ia laikui b gant. Galima sp sti, kad  i priel
%K spriklers
%K fire
%K Bayesain approach
%K multi-attribute selection
%U http://www.mla.vgtu.lt/index.php/mla/article/view/186