Getting Alarms Right the First Time or Fixing Later

As distributed control systems, human-machine interface (HMI)-based control systems and safety instrumented systems came along, adding alarms was easy and close to free. The problem they introduced was too many too often—whether immediate action was warranted or not. The ISA 18.2 alarm management standard was developed to help manufacturers and producers optimize the alarms to minimize alarm flooding conditions and provide clear notification when action was required.

Emerson's Matt Stoner


I bring this up because there is a thoughtful conversation going on in a LinkedIn status update I shared this morning. This update featured Emerson’s Matt Stoner and a short video where he describes good and bad alarms and where a plant might be with respect to the ISA 18.2 standard.

Here’s the short minute and a half video, Cut through the Noise: Creating a Better Alarm System.

A comment came in from an oil & gas industry veteran with long experience in instrumentation and controls:

This may be radical but I have often wondered why alarm management is even necessary. Project Teams create the alarms during the design of the facility and then there is a major effort to manage the alarms. That seems to be a very inefficient process. Why couldn’t there be proper templates and guides so that when the alarms are originally created, they are proper and meet all of the industry and owner practices so that alarm management is not necessary. When the facility design is complete, the alarm system is proper and does not have to be managed. It just seems all industries have accepted the historical processes which involves creating an alarm volume or flood problem and then correcting it later.

He added:

I have one additional thought. I believe that most of the tools that have been created for alarm management were developed specifically for existing facilities that already have alarm problems. I believe the tools assume an alarm problem already exists. Today, we have to find every single opportunity to stop and reverse the historical trend of escalating project costs. Allowing alarms to be created like they have for decades and then using tools primarily intended for operating facilities to manage alarms for new facilities (i.e. rework) seems inefficient and costly. Can there be different tools, guides and templates for greenfield projects that eliminates the rework process? Can we get it right when the alarms are created so that no management or rework is necessary?

I shared these thoughts with Matt and he responded back:

Many years ago, I was told that alarms use to “cost” (assume $) on control systems so alarm creation was limited. Sometime later these alarms became free and this allowed the process engineers, project teams and/or plant managers the ability to add whatever alarms they thought ‘might’ be needed by ‘someone’ without any focus on the result of these additions to the person running the board. I have seen alarms that were defaulted to be enabled on every point in some systems. All the projects that I have been involved with recently have some form of alarm rationalization being done (new and migrations). We have been pushing the ISA 18.2 alarm management process and specifically making anyone asking for an alarm to have it follow an alarm rationalization process and/or fill out the DeltaV Alarm Help items (recommended action, time to respond, consequence of inaction, probably cause and design information). This really helps identify if it really should be an alarm for the person running the board or if it should be something different.

I agree with what you are saying and I believe that most are starting to or already doing that in today’s project executions…it just doesn’t mean that projects did it 15, 10 or even 5 years ago when they initially configured the system and now something must be done. There have been some good presentations at Exchange the last couple years on people going thru the alarm management process. They indicated a painful and very difficult process because of logistics of getting everyone needed together in a running facility but all of them seemed to have seen value after completing just portions of their facilities and now have processes in place to keep it from occurring again.

If you have observations or additional points to make, share them in the Emerson Exchange 365 DeltaV group discussion thread, VIDEO: Cutting through the Noise to Find the Best Alarm Management System.

Posted Thursday, December 7th, 2017 under Alarm Management.