Alarm Shelving Priority Engine for Nuisance Alarm Floods
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for setting alarm limits in industrial processes are not scientifically based, leading to issues like alarm floods, standing alarms, and nuisance alarms, which can result in inefficient operation and potential accidents due to operators ignoring important alarms.
Innovation Solution
A decision support system that uses a shelving decision engine to analyze industrial process information and historical data to determine the priority of alarms, automatically shelving them based on a weighting algorithm that considers relevant factors, thereby assisting operators in managing nuisance alarms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If alarm shelving is performed manually by operators, then operators can control which alarms to shelve, but operator workload increases and decision accuracy decreases due to alarm floods and nuisance alarms
Solution Approach 1:
The system enables alarms to be shelved automatically based on their own characteristics and historical data, without requiring operator intervention. The shelving decision support engine analyzes alarm properties, correlates with historical information, and autonomously determines shelving actions, allowing the system to serve itself in managing nuisance alarms.
Solution Approach 2:
The shelving decision support engine acts as an intermediary between the alarm system and operators. It processes alarm information, applies weighting algorithms, and generates shelving recommendations, thereby mediating the complex decision-making process and reducing the cognitive burden on operators while improving shelving accuracy.
2Loss of information
If more alarms are displayed to operators, then complete information is provided, but operator attention is分散ed and important alarms are missed due to alarm floods
Solution Approach 1:
The system extracts and removes nuisance alarms from the operator's view by automatically shelving them based on analysis of alarm characteristics and historical data. This separation allows operators to focus on critical alarms without being distracted by redundant information, while the shelved alarms remain stored and can be retrieved when needed.
Solution Approach 2:
The alarm system is segmented into different priority levels, with the shelving decision support engine categorizing alarms based on their importance and characteristics. Critical alarms are presented to operators while less important alarms are shelved separately, creating a hierarchical structure that improves information management and operator focus.
3Ease of operation
If automatic shelving is implemented, then operator workload is reduced, but shelving accuracy may decrease without proper decision support
Solution Approach 1:
The shelving decision support engine incorporates feedback mechanisms that continuously learn from operator actions and alarm outcomes. By analyzing historical shelving decisions and their results, the system refines its weighting algorithms and improves the accuracy of automatic shelving recommendations over time, creating a closed-loop learning system.
Solution Approach 2:
The system dynamically adjusts the parameters and weights in its decision-making algorithm based on historical data and changing operational conditions. By modifying the weighting factors assigned to different alarm characteristics, the system adapts to improve shelving accuracy while maintaining automated operation, balancing ease of use with decision precision.
Data Source
AI summary
System and method for determining a probability of an alarm being shelved using a weighting algorithm taking into account relevant factors or characteristics associated with an industrial process. Support for rendering a decision is provided, including determining an estimated time duration before the alarm is un-shelved if the alarm is likely to be shelved, a rationale for the alarm to be shelved, and relevant information supporting the rationale. A machine learning model learns from operator actions relating to alarms to capture the knowledge of the operator and to update or refine the weighting algorithm.