The Design of a Fuzzy Logical Vigilance Performance Alarm System for Automation Supervisory Control Tasks

Abstract

Previous researches have shown that the vigilance determines the reliability of human performance, especially in low arousal level tasks such as the radar monitoring, so it is an important factor for automation system performance. The purpose of this research was to study how the fuzzy technique can combat the loss of vigilance in automation. Firstly, a quantitative performance measuring model with signal to noise ratio (η value) for dynamic characteristics of human decision-making was proposed. Secondly, using the terms of the performance evaluating value (η value) and the stochastic process in automation signal status, a fuzzy logical vigilance performance alarm system was constructed. The results of this study indicated that using the performance measuring model to measure system operating performance on human dynamic decision-making is easy and objective, and the fuzzy logical vigilance performance alarm combined with financial incentives according to the η value could improve the system operating performance significantly.