Analysis of false alarms given by automatic fire detection systems

Abstract

The high frequency of false alarms sounded by automatic fire detection (AFD) systems can result in substantial monetary losses and increased risk to human life. In this paper, we present an analysis of false alarm data, which were collected from six different sites, to estimate the time interval in which the next false alarm will occur. This analysis is performed on the basis of the Poisson and the non-homogeneous Poisson process.