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

Douglas Rothenberg, Ph.D.

A functional alarm system is one of the cornerstones for responsible operation of industrial plants.  Done properly it will enable the operator to be notified of abnormal situations in time to make the difference.  An alarm activation is the last chance a plant or enterprise will have to place the operator in the solution/resolution path before the abnormal situation runs its course.  If proper safeguards like interlock shutdowns or overpressure relief system and the like are there and working, they will provide a measure of eventual protection.  Operator action before then is preferred.  To have the proper alarms, it is essential that equipment designers and operations experts identify in advance all reasonable abnormal situations and provide appropriate alarms for them.

What to Alarm

The figure below illustrates the operational situation for designing alarms:

Rothenberg Newsletter Pic

Lest we miss the point, it is a clear responsibility to deep dive into the plant and find all of the reasonable abnormal situations.  Each one needs to be properly alarmed through the alarm rationalization process.  This work process is nicely structured in a way that is efficient and promotes predictably and an appropriate degree of uniformity. But wait a minute!  What you are going to read next is fundamental alarm management but may be light years away from what might be thought of as “good” rationalization.  During the past several years, we’ve come to a really better understanding of how to do rationalization.  No longer is it even useful, forget being the way to do it, to work down the list of existing or potential alarms and decide “yes” or “no” for each.  Now there is an entirely different way.  We’re talking about the only really effective methodology.  How else can we make sure that all reasonable abnormal situations are exposed and managed?  To drive this point home, consider this situation: you’ve decided that a certain specific temperature needs to be kept below a trouble point limit to avoid overheating the contents of a mixer.  You alarm it enough before the trouble point for the operator to fully work through the alarm make the needed adjustments so the contents of mixer avoids getting too hot.  But, without a full analysis of all of the abnormal situations for which the temperature of the mixer contents needs to be managed, it would be entirely possible to miss the situation where a downstream separator must work at an even lower temperature.  You can think of much better examples of this I’m sure.

Step by Step

First and foremost, make sure the team doing the alarm rationalization is very experienced and together have sufficient knowledge of the entire plant and its operation.  In addition to operators, you’ll need equipment specialists, operation support engineers, and safety and financial expertise.  At the end of the day, the enterprise will end up with a fuller and more grounded understanding of all operations. Let’s take this alarm rationalization process step-by-step.

Before Starting

Identify all “required” alarms.  Required alarms are such that regardless of any other basic alarm management guidelines must be present due to legal requirements, statuary requirements, or enterprise demands.  All will be properly rationalized (step 1 through step 5), even if some or all would not be actually have been selected to be alarms according to good alarm management practices.

Step 1

Identify all individual key components of a plant: e.g., pumps, blowers, compressors, heat exchangers, reactors, crushers, separators, storage systems, utility systems, and the like.  This list should contain 90 plus percent of the parts of the plant.  If appropriate, the list is further subdivided into large ones, small ones, steam driven or electrical drives, and such.

Step 2

For each individual key component (one at a time) define the purpose or job of it.  For example, the “job” of a heat exchanger is to move thermal energy between a process fluid and an energy fluid.  Thoughtfully identify them all.

Step 3

Now taking each key component ask what process condition, operational condition, or issue can get in the way of the selected key component doing its job?  For the heat exchanger example, only two things can get in the way: not enough thermal driving force between the process fluid and the energy fluid; and/or not enough flow through the exchanger (generally an unacceptably high resistance to fluid flow in the tubes for a shell and tube exchanger) to enable sufficient energy transfer.

Step 4

Taking each condition or issue above one-by-one, ask what is the best thing to alarm so that the operator will know there is a problem?  For our heat exchanger example, there would be a too-low temperature difference between the two fluids; and there would be a too-high pressure difference across the tubes.

Step 5

For each alarm identified above, produce the full set of rationalization and configuration information: causes, confirmatory actions, manual corrective actions, consequences, advanced alarm controls, testing, urgency, priority, alarm activation point, and any other needed documentation.

Step 6

Repeat step 2 through step 5 for all the key components of the plant or enterprise.  Now you should have a fully rationalized alarm set for each of the key components.  These will be used as starting-point templates for those key components everywhere they occur in your plant or enterprise.

Step 7

Work through your entire plant by picking up the appropriate alarm template set for the key components as the starting point.  Then ‘adjust’ the template for what might be differences between the template and what exactly is there.

Finally

Take a step back and look for anything missing.  Examine the entire set of old existing alarms to see if anything in the new alarm set might have been missed or needs a re-visit.  You are looking for the odd missing abnormal situation.  This is not a back door for having lots of old alarms back.

Tool Set

Rationalization gets done by teams who do the hard work.  Understand the process.  ANSI/ISA-18.2, API RP-1167, EEMUA 191, and Alarm Management for Process Control are very helpful references and guides.  Using a purpose built tool like LogMate® will facilitate the rationalization process, link up with the alarm configuration databases of the control systems, and directly export to on-line documentation files for alarm response.