The short answer
Alarm Priority
Alarm priority tells the operator which alarm to act on first. It is assigned from two things: the severity of the consequence if no action is taken, and the time available to act. A small matrix of those two produces three or four priorities. ISA-18.2 and EEMUA 191 recommend a distribution of roughly 80 percent low, 15 percent medium, and 5 percent high, so that a high-priority alarm is rare and always means something.
Key points
- Priority is about the consequence of not acting and the time to act, not about how important the equipment feels.
- Three priorities, or four with a critical level, is all an operator can use.
- Aim for about 80/15/5 low/medium/high across the configured alarms.
- Priority is set once, in rationalization, with the consequence written down.
- If everything is high, nothing is.
What priority is for
When two alarms arrive at once, the operator needs to know which to act on first. That is the only job of alarm priority. It is not a measure of how expensive the equipment is, how loudly a supervisor argued for it, or how embarrassing the failure would be. It is a ranking of urgency, and it only works if it is assigned consistently across every alarm in the system by a method that can be explained.
Priority also drives presentation. The alarm summary sorts by it, the HMI color and sound depend on it, and in a utility with paging or callout, priority is often one of the inputs that decides whether an alarm wakes someone at 2 a.m. or waits for the morning shift. Getting priority wrong in either direction has consequences: a low priority on a chlorine leak is obviously wrong, but a high priority on a communication hiccup trains operators to ignore high priority alarms.
The two questions
ISA-18.2 defines priority as a function of the severity of consequence and the time available for response. Both are judged assuming the operator does nothing: what happens, how bad is it, and how long until it happens.
| Time to respond | Minor consequence | Major consequence | Severe consequence |
|---|---|---|---|
| More than 30 minutes | Low | Low | Medium |
| 5 to 30 minutes | Low | Medium | High |
| Less than 5 minutes | Medium | High | High |
The matrix itself is chosen by the utility, and the words minor, major, and severe are defined in the alarm philosophy against categories that matter to the site: personnel safety, public health, environmental release, regulatory violation, equipment damage, and cost. A minor consequence might be an equipment repair under a set amount or a brief loss of a non-critical measurement. A severe one is an injury, a sanitary sewer overflow, a treatment violation, or a loss of the ability to supply water. Once the definitions exist, two people rationalizing the same alarm arrive at the same priority, which is the test.
How many levels
Three is the common answer: low, medium, high. Some sites add a fourth, critical or emergency, reserved for a handful of alarms with an immediate safety consequence: a gas detector, a fire alarm, an emergency stop. Beyond four, operators cannot keep the distinctions straight, and the extra levels get used as a way to avoid deciding.
Diagnostic and maintenance notifications that do not require operator action are not alarms at all under ISA-18.2, and should not occupy a priority. Route them to a maintenance list or an event log. A lot of the pressure to add priority levels comes from trying to fit things that are not alarms into the alarm system.
Distribution
The published guidance from EEMUA 191, carried into ISA-18.2, is a target distribution of about 80 percent low, 15 percent medium, and 5 percent high priority across the configured alarms, with under 1 percent in a critical level where one exists. The numbers are not magic. They express the idea that a high-priority alarm should be uncommon enough that it commands attention every time.
A system where 40 percent of alarms are high has not been rationalized; it has been configured by whoever was most worried at the time. The fix is not to demote alarms arbitrarily until the numbers fit but to run each one through the matrix. The distribution then falls out, and if it is still top-heavy the consequence definitions probably need tightening.
| Priority | Target share | Operator expectation | Example at a water utility |
|---|---|---|---|
| Critical (optional) | Under 1% | Drop everything; immediate action, possibly evacuate | Chlorine gas detection, fire, emergency stop actuated |
| High | About 5% | Act now, ahead of anything else on the screen | Lift station high level, disinfection residual out of range at the point of entry, plant loss of power |
| Medium | About 15% | Act within minutes, after any high | Pump failed to start with the standby available, tank level approaching low, communication lost to a remote site |
| Low | About 80% | Act when convenient within the shift | Filter run time due for backwash, analyzer calibration due, a single flow meter reading suspect |
Common mistakes
- Priority by equipment
- Everything on the high-service pumps is high because the pumps are important. The pumps are important; the alarm about a bearing temperature rising slowly is not urgent.
- Priority by who asked
- A supervisor wants to be sure the alarm is noticed, so it is set high. Priority is the answer to the matrix, and the way to be sure an alarm is noticed is to have few high alarms.
- Priority to control notification
- An alarm is set high so it pages someone. If the notification system needs a separate criterion, give it one. Coupling the two forces the wrong priority on alarms that need a page but are not urgent, and the reverse.
- Same priority for the pre-alarm and the alarm
- High level and high-high level on a tank are two alarms with different times to respond. The first is medium, the second is high. Giving both high makes the first one noise.
- Priority never revisited
- A priority assigned at commissioning under commissioning conditions stays for twenty years. Rationalization is periodic, and the alarm history shows which priorities are wrong.
Priority in presentation
Once assigned, priority should be visible the same way everywhere. ISA-101 and the high-performance HMI practice give each priority a distinct color and, where used, a distinct sound, and reserve those colors for alarms only. The alarm summary sorts by priority and then by time. The alarm indication on a process display uses the priority color and shape. A pager or callout system uses the priority as one input to its escalation rules. When priority looks different on different screens, the operator has to translate, and translation under stress fails.
Frequently asked questions
- Should communication failure alarms be high priority?
- Usually medium. Losing communication to a lift station means losing visibility, which needs attention within minutes, but the station keeps pumping on its own controls and the consequence of a short outage is small. It becomes high if the site cannot run without SCADA, which is a design problem to fix rather than a priority to raise.
- Can the same alarm have different priorities at different times?
- ISA-18.2 allows state-based priority, where the priority depends on the operating mode. It is useful in batch and process plants and rarely needed in water and wastewater. If the same alarm seems to need two priorities, it is often two alarms, or the time to respond has been misjudged.
- What priority do I give an alarm I cannot classify?
- Ask what happens if nobody responds for an hour. If the honest answer is nothing much, it is low, or it is not an alarm and belongs in the event log. Inability to name a consequence is itself the answer.
- How do I fix a system where most alarms are already high?
- Start with the alarm history. The most frequent alarms are the first candidates, because they are the ones operators have learned to ignore. Rationalize those, then work through the rest by area. Reducing the high priority count to the target takes months at most sites and is worth it every time.
Related topics
- ISA-18.2 Alarm ManagementThe lifecycle standard for alarm systems, the rate targets that define a workable system, and why alarm rationalization is the step nobody wants to fund.
- The Alarm PhilosophyThe document that decides what is allowed to be an alarm: the definition, the criteria, the priorities and their meaning, the performance targets, the handling rules, and who owns it. Under ISA-18.2 it comes first.
- Alarm RationalizationThe meeting where every alarm earns its place: who attends, what is decided for each alarm, how to document it in a master alarm database, and how to run it at a utility that cannot spare a week.
- Alarm FloodsWhat an alarm flood is, why it happens at exactly the wrong moment, the ISA-18.2 rate targets, and the design measures that keep a power failure or a communication loss from burying the one alarm that matters.
- ISA-101: Human Machine InterfacesWhat the ISA-101 standard for human machine interfaces covers, its lifecycle from philosophy to maintenance, the display hierarchy, and what adopting it changes on a project.
- Alarm ShelvingThe operator tool for temporarily silencing a nuisance alarm: what shelving is and is not, the rules a philosophy sets, who may shelve what and for how long, automatic unshelving, the shelved list, the record, and how to implement it on any platform.
Direct contact
Have a controls question?
Reach Eric Sullivan directly about anything on this site, a controls or automation topic, or one of his personal projects.