The Analyses of the Reliability of Building Fire Protection Systems
Xing MAO,Mao LIU
DOI: https://doi.org/10.3969/j.issn.1009-6094.2008.01.042
2008-01-01
Applied Mechanics and Materials
Abstract:The present paper is inclined to introduce a data-driven method and based on the Bayes' Theory and Monte Carlo simulation. In doing so, the first step we would like to introduce is the degree of complexity of the fire protection system. In our research, we divide the building fire protection system into essential components in a form of an accident- or fault-tree. Next, different response scenarios is designed corresponding to the likely or unexpected accidents, which can be illustrated and analyzed by using the last step, that is, the presupposed moment when the fire accident occurs, by means of the event tree. After that, we have to apply the Bayes' Theory to disposal of the historical statistic data. In this way, it is expected for us to receive the uncertainty probability distribution of the system failure rate so as to establish the function of the system reliability, which is up to the time function with the help of the reliability analysis. And, now, Monte Carlo simulation can be adopted to analyze the uncertainty of the reliability, the time function to estimation the system reliability. And, finally, Sensitivity Analysis can be done, which is expected to inform us of the degree of influence of different uncertainty variables on the total uncertainty. Thus, integrating the statistic data, empirical formula, and Bayes' Theory, will enable us to estimate the reliability of the building fire protection system much more effectively. The suggested method of ours can be used to evaluate the reliability of other similar systems. In addition, it can also help us to reduce the uncertainty by Sensitivity Analyses so as to improve the accuracy of the estimation.