Why Choose TiPS?

LogMate is the most reliable, proven software solution since 1990

    • Leadership Participation in Alarm Standards Committees
    • Dedicated Alarm Management Focus
    • Comprehensive Alarm Management Software Solutions
    • Adaptable Services Model
    • Flexible and Extensive Software Connectivity Portfolio
    • Broad Industry Experience

Introduction to Alarm Management

Mike Lyssy

 

Facilities tackle the issue of process complexity and suboptimal or nuisance alarm count by conducting an alarm rationalization.

 Once this resource intensive effort is completed, can the rationalization team relax, congratulate each other on a job well done and go back to the routine operation tasks in the facility?

A Big No! Of Course No! The problem is as time passes and change comes to the process, performance of the alarm system is not tracked. Maintaining improvements like optimized alarm load and incident reduction through timely advisory Information can be difficult if you don’t efficiently follow up a rationalization with the proper continuous improvement program. This article is the first in a series describing the steps a diligent engineer should conduct once a rationalization has been completed. The focus for this article will be Alarm Performance Analysis.

The Opera Ain’t Over

In an evergreen alarm management process, the automation professional keeps asking, “Why?” and “Why not?” Prudent automation personnel will expect degradation in performance and should take on the task of insuring this work product does not deteriorate. So, after completion of the alarm rationalization operations should ask questions about performance of the alarm management process.

    • • How well did we do in the alarm rationalization?
    • • Did we look at alarm performance prior to the alarm rationalization so we have benchmark KPI’s to give an indication of improvement?
    • • Did we follow the site’s alarm philosophy?

 

Attack: Summary of the Steps to Take

First the assumptions:

        • • A reasonable alarm philosophy is in place that includes directions on maintaining the alarm system.
        • • A benchmark of performance was taken prior to the alarm rationalization to establish a baseline.

 
Taking into account the assumptions, the automation professional should consider the following:

1- Analyze Alarm Performance Periodically

2- Proactive Management for Successful Change

3- Industrial CyberHouse Considerations and Team Accountability

 

Analyze Alarm Performance Periodically

For best practice standard metrics and more advanced analysis, both static (alarm configuration) and dynamic (alarm occurrence) data are critical to good analysis of alarm management performance.

Static Data and Metrics

The understanding of the “what and how” of good alarm limit settings (often called trip points), priority distribution, enable status, dead bands and any associated displays is contained in static data. Ideally you have archived information downloaded from the control system alarm configuration files and information developed during the rationalization in a master alarm database (MAdB). Static metrics help establish desired performance of the alarm system.

Table 1 Static Metrics

Metric How it is done Indications
Total number of configured alarms
• BPCS dump of configured Tags
• Report from Alarm Rationalization tool
• Export from Alarm Management software solution

 

Too many configured alarms can set the facility up for potential alarm floods or nuisance or non-essential alarming during an event.
Average number of alarm points per tag
Compare the number of configured alarms to the total number of tags that have alarms configured
Example: Alarms per Controller:

  • • Low – 1
  • • Average – 4
  • • High – 6
Use of different alarm types
• Sort from BPCS dump of configured Tags
• Report from Alarm Rationalization tool
• Export from Alarm Management software solution
Certain alarm types can be a source of nuisance in various process conditions, such as deviation, rate of change, etc.
Use of BADPV, etc.
• Check rule template from the alarm rationalization
• Sort from BPCS dump of configured Tags
Bad Measurement type alarms come in an array of configurations. These can become a source of nuisance as well as cause load on the event log historian.
Spread between trip points
• Sort from BPCS dump of configured Tags
• High and high-high trip points to close can cause nuisance alarms and not allow enough operator reaction time
• Over use of trip points at both ends of the span
Alarm priority distribution
• Pareto Chart from Alarm Rationalization tool
• Export from Alarm Management software solution
• High – 5% of Total
• Medium – 15% of Total
• Low – 80% of Total

Dynamic Data and Metrics

Reading the key performance indicators derived from dynamic data will keep you on course. Dynamic data includes the activation data: time, message, alarm type, alarm priority presented to the operator; information found in the control systems alarm and event data string. Access to dynamic data will support the benchmarking documented in best practice recommendations and before and after the rationalization to measure any effectiveness. You can find metrics for performance monitoring in best practice and recommended practices publications (e.g. ISA 18.2, API RP 1167.) Below you will find common dynamic metrics found in most of the alarm management best practice standards.

Table 2 Common Alarm Management Key Performance Indicators

Metric

Annunciated Alarms Per Time
Annunciated Alarms Per Day per controller Position
Annunciated Alarms Per Hour per controller Position
Annunciated Alarms per 10 Minutes per Controller Position
Percentage of Hours containing more than 30 alarms
Percentage of 10-minute periods containing more than 10 alarms
Maximum number of alarms in a 10 minute period
Percentage of time the alarm system is in a flood condition
Percentage contribution of top 10 most frequent alarms to the overall alarm load
Quantity of chattering and fleeting alarms
Stale Alarms
Annunciated or Configured Priority Distribution
Unauthorized Alarm Suppression
Improper Alarm Attribute Change

With these configuration metrics we can see if we have reduced the potential for load on the operator by lessening the number of alarms configured.

Continue The Cycle

Rationalization doesn’t stop after the workshop is completed. Alarm management is an evergreen process, and as such requires a proper continuous improvement program by committed and diligent automation personnel. Don’t let your rationalization team’s hard work go to waste. To request more information on maintaining the benefits of your rationalization, click here.

 

Haven’t rationalized yet?Click here for more information.