Industrial Plant Alarm Prioritization Using Causal State Filtering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Industrial plants face challenges in determining the operational state and identifying critical alarms amidst a flood of notifications, making it difficult for operators to take corrective action effectively due to the noise of irrelevant or secondary alarms.

Innovation Solution

A computer-implemented method filters alarms based on causal chains and state-dependent conditions to prioritize service alarms over process alarms, moving non-essential alarms to an informative pool, allowing operators to focus on immediate corrective actions that prevent process halts.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If all alarms are displayed to operators, then complete information is provided, but the signal-to-noise ratio deteriorates making it difficult to identify critical alarms

Engineering Contradiction:
Improvecomplete alarm informationVSAvoidoperator ability to identify critical alarms
Core Design Contradiction:
Loss of informationVSEase of operation

Solution Approach 1:

The alarm system segments alarms into different categories (service alarms and process alarms) and displays them separately. Service alarms that indicate potential service disruptions are prioritized and displayed first, while process alarms are displayed separately or suppressed. This segmentation allows operators to focus on critical alarms without being overwhelmed by less important notifications, resolving the contradiction between providing complete information and maintaining ease of operation.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system introduces an intermediary filtering mechanism that processes alarms before presenting them to operators. This intermediary layer analyzes alarm characteristics, determines their priority and category, and selectively displays or suppresses alarms based on predefined criteria. This mediator resolves the contradiction by automatically filtering the alarm flood while preserving all information for later review, making critical alarms easily identifiable without losing complete information.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If all alarms are displayed, then comprehensive monitoring is achieved, but operator response time deteriorates due to alarm flood

Engineering Contradiction:
Improvecomprehensive plant monitoringVSAvoidoperator response time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

By segmenting alarms into service alarms (critical) and process alarms (less critical), the system enables operators to respond to service alarms immediately while knowing that process alarms are being monitored separately. This segmentation maintains comprehensive monitoring coverage while reducing the time operators need to respond to each individual alarm, as they can prioritize based on the segmented categories.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary classification and prioritization of alarms automatically before they reach the operator. By pre-processing alarms to identify service alarms that indicate potential service disruptions, the system prepares the information in advance, allowing operators to respond immediately to critical issues without spending time analyzing the full alarm list, thus reducing response time while maintaining comprehensive monitoring.

Inventive Principle:
Principle #10Preliminary action

3Loss of information

If process alarms are displayed, then detailed process information is provided, but service alarm detection deteriorates due to noise from less critical alarms

Engineering Contradiction:
Improveprocess state informationVSAvoidservice alarm detection
Core Design Contradiction:
Loss of informationVSDifficulty of detecting and measuring

Solution Approach 1:

The system segments the alarm display into separate sections for service alarms and process alarms. Service alarms are displayed prominently with higher visibility, while process alarms are displayed in a separate area or with lower prominence. This segmentation ensures that detailed process information is preserved and available, but service alarms are easily detectable without being obscured by process alarm noise.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system applies different display qualities and priorities to different alarm types. Service alarms receive higher local quality in terms of display prominence, alertness, and priority treatment, while process alarms are displayed with standard or reduced quality. This local differentiation ensures that service alarm detection is not difficult despite the presence of process alarms, while still providing detailed process information.

Inventive Principle:
Principle #3Local quality

Data Source

PatentEP4111278B1A computer-implemented method for determining an operational state of an industrial plant
Publication Date: 2024.02.14 ABB (SCHWEIZ) AG
  • EP4111278B1 patent drawingFigure 1
  • EP4111278B1 patent drawingFigure 2~3

AI summary

A computer-implemented method (100) for determining an operational state of an industrial plant (1) executing an industrial process, the method (100) comprising: • adding alarms (2) raised within the plant (1) to a first pool (3a) of important alarms (2); • determining (120), based on plant (1) topology, on the process, and/or on a physical state (1c) of parts of the plant (1), whether a physical state (1c) indicated by a first acquired alarm (2) has caused a second alarm (21), and if so, moving (130) the first alarm (2) from the first pool (3a) to a second pool (3b) of informative alarms (2); and/or • determining (140), based on the topology of the plant (1), and/or on the process, in combination with a physical state (1c) of parts of the plant (1), for at least one first acquired alarm (2), whether a predetermined state-dependent condition (5) is met, and if this is the case, moving (150) the first alarm (2) from the first pool (3a) to the second pool (3b); and • determining (160) the operational state (la) of the plant (1), and/ or a corrective action (lb) based on the alarms (2) in the first pool (3a).