Registers
Alarm List and Rationalisation Record
ALMISA-18.2 asks a hard question of every alarm: what action does the operator take, and what happens if they do not take it in time? An alarm with no answer is not an alarm. Working through this document is what prevents the alarm flood that makes a control room ignore the horn.
Do you need this one?
Any system with an operator interface. On a small machine the list is short, but the discipline still applies.
What is in it
- 01Alarm identity and trigger
- 02Priority with justification
- 03Consequence of inaction
- 04Time to respond
- 05Corrective action
- 06Rationalisation sign-off
Written by
Control engineer with the operations team
Approved by
Operations manager
Written against
- ISA-18.2 / IEC 62682
Preview
Alarm Tag,Message Text,Source Tag,Trigger Condition,Priority,Consequence of Inaction,Time to Respond,Operator Action,Class,Deadband,On Delay s,Off Delay s,Auto Ack,Suppressed When,Rationalised By,Date AL-001,Emergency stop activated,ESTOP-01,Signal de-energised,High,Plant remains stopped,Immediate,Identify station and make safe before reset,Safety,,0,0,No,,, AL-002,Guard door open,GS-101,Signal de-energised,High,Machine cannot run,Immediate,Close and latch guard then reset,Safety,,0,0,No,Maintenance mode,, AL-003,Feed tank level high high,LT-101,> 95%,High,Overflow and spill,1 minute,Check outfeed route and clear blockage,Process,2%,2,5,No,,, AL-004,Feed tank level low low,LT-101,< 5%,High,Pump runs dry and is damaged,1 minute,Stop pump and restore feed,Equipment,2%,5,5,No,Pump stopped,, AL-005,Reactor temperature high,TT-101,> 150 degC,Medium,Off specification product,10 minutes,Reduce feed rate and check cooling,Quality,2 degC,0,5,No,,, AL-006,Reactor temperature high high,TT-101,> 180 degC,High,Runaway reaction,Immediate,Confirm master trip and investigate,Safety,2 degC,0,0,No,,, AL-007,Feed flow low,FT-101,< 5 m3/h,Medium,Sequence stalls,5 minutes,Check strainer and valve position,Process,1 m3/h,10,10,No,Pump stopped,, AL-008,Feed pump fault,MTR-101,Starter aux open with run command,Medium,Loss of feed,5 minutes,Check overload and reset at MCC,Equipment,,3,0,No,,, AL-009,Valve failed to open,XV-101,No open limit 10 s after command,Medium,Sequence holds,5 minutes,Check air supply and limit switch,Equipment,,10,0,No,,, AL-010,Remote I/O communication loss,,Watchdog expired,High,Loss of control and indication,Immediate,Check network and panel power,System,,3,0,No,,, AL-011,Control power failure,,24VDC below threshold,High,Total loss of control,Immediate,Restore supply then restart per procedure,System,,0,0,No,,,
# Alarm Rationalisation Record
## {{PROJECT_NAME}}
| | |
|---|---|
| **Document title** | Alarm Rationalisation Record (ALM) |
| **Document number** | {{DOC_NO}} |
| **Revision** | {{REV}} |
| **Date** | {{DATE}} |
| **Project** | {{PROJECT_NAME}} |
| **Client** | {{CLIENT}} |
| **Prepared by** | {{AUTHOR}}, {{COMPANY}} |
### Revision history
| Rev | Date | Description | Author | Approved |
|---|---|---|---|---|
| 0 | {{DATE}} | First issue | {{AUTHOR}} | |
| | | | | |
### Approval
| Role | Name | Signature | Date |
|---|---|---|---|
| Author | {{AUTHOR}} | | |
| Reviewer (engineering) | | | |
| Approver ({{COMPANY}}) | | | |
| Approver ({{CLIENT}}) | | | |
---
## Why this document exists
An alarm exists to tell an operator to do something they would not otherwise do, in time for it
to matter. ISA-18.2 makes that testable with three questions, and every alarm on the list must
answer all three:
1. What is the **consequence** if the operator does nothing?
2. How long do they have to **respond** before that consequence occurs?
3. What **action** do they take?
If a proposed alarm has no answer to question 3, it is an indication and belongs on a status
display. If it has no answer to question 1, delete it.
## Priority assignment
Priority follows from consequence severity and from the time available, not from how important
the equipment feels.
| Priority | Consequence | Time to respond | Target share of alarms |
|---|---|---|---|
| High | Injury, environmental release, major loss | Under 3 minutes | 5% |
| Medium | Product loss, equipment damage | 3 to 30 minutes | 15% |
| Low | Efficiency, minor quality | Over 30 minutes | 80% |
## Performance targets
From ISA-18.2, per operator console:
| Metric | Target | Maximum acceptable |
|---|---|---|
| Average alarms per hour | 6 | 12 |
| Peak alarms in 10 minutes | 5 | 10 |
| Time in flood condition | 0% | 1% |
| Standing alarms | 0 | 5 |
Measure these during the SAT and again after 30 days of operation. An alarm system that meets
them on day one and not on day thirty has a chattering alarm somewhere, and the deadband or the
delay is wrong.
## Rationalisation session record
| Date | Attendees | Alarms reviewed | Added | Removed | Re-prioritised |
|---|---|---|---|---|---|
| | | | | | |
## Chattering and stale alarms
| Alarm | Occurrences per hour | Cause | Action | Closed |
|---|---|---|---|---|
| | | | [Deadband, delay, or fix the instrument] | |