A Systematic Approach to Design and Analysis of Univariate Alarm Systems Using Penalty Approaches

This paper presents a new method for designing a generalized delay timer based on penalty scenario and Markov chain schemes. The penalty approach is the extension of n-sample on/off delay approaches in designing alarm systems. Three performance indices named, False Alarm Rate (FAR), Missed Alarm Rate (MAR) and Average Alarm Delay (AAD) are derived for penalty approaches using Markov models. Also, the new index named “Mean Time to Alarm (MTTA)” is introduced to analysis alarm system and computation of AAD in penalty scenario. The optimal values of delays and penalties are derived based on target function using Genetic Algorithm (GA). Finally, the effectiveness of the proposed method is investigated and compared with other methods through a case study.