Dissonance between multiple alerting systems.I. Modeling and analysis

The potential for conflicting information to be transmitted by different automated alerting systems is growing as these systems become more pervasive in process operations. Newly introduced alerting systems must be carefully designed to minimize the potential for and impact of alerting conflicts, but little is currently available to aid this process. A model of alert dissonance is developed that provides a theoretical foundation for understanding conflicts and a practical basis from which specific problems can be addressed. Part I establishes a generalized methodology to analyze dissonance between alerting systems, and Part II exercises the principles and presents methodologies to avoid and mitigate dissonance. In Part I, we develop a generalized state-space representation of alerting operation that can be tailored across a variety of applications. Based on the representation, two major causes of dissonance are identified: logic differences and sensor error. Additionally, several possible types of dissonance are identified. A mathematical analysis method is developed to identify the conditions that cause dissonance due to logic differences. A probabilistic analysis methodology is also developed to estimate the probability of dissonance originating due to sensor error.