An Adaptive Optimization Approach Based on the Human factors and Its Application to Process Alarm Thresholds

Currently, the majority of studies on process alarm thresholds optimizations are less concerned with human factors. In response to this limitation, a novel adaptive optimization approach to process alarm thresholds concerning human factors is explicitly introduced in this paper. Apriori algorithms a...

Full description

Saved in:
Bibliographic Details
Published in2019 12th Asian Control Conference (ASCC) pp. 823 - 828
Main Authors Li, Jince, Wang, Yongjian, Li, Hongguang, Yang, Bo
Format Conference Proceeding
LanguageEnglish
Published JSME 01.06.2019
Subjects
Alarm systems
Fuzzy reasoning
Human factors
Indexes
Itemsets
Mathematical model
Optimization
Online AccessGet full text

Cover

More Information
Summary:Currently, the majority of studies on process alarm thresholds optimizations are less concerned with human factors. In response to this limitation, a novel adaptive optimization approach to process alarm thresholds concerning human factors is explicitly introduced in this paper. Apriori algorithms are used to extract association rules between process alarms and manipulated variables. By means of fuzzy reasoning strategies, a human factor index in terms of operators' perception and treatment capabilities is created before the nuisance alarm rate and alarm treatment rate are calculated, respectively. Subsequently, an thresholds optimization objective function with respect to the weighted false alarm rate (FAR) and missed alarm rate (MAR) is established, in which, the FAR and MAR are estimated with kernel density methods while weights are calculated based on the human factor index. Case studies show that the proposed method can adapt to operators' human factors, facilitate the process alarm treatment and improve the process safety.