Horrible Things We Find in Alarm Rationalization

  • Conference Program
  • Hydrocarbon Processing
  • April 10, 2019
  • 9:15 am - 10:00 am

Alarm Rationalization is the most difficult, resource-intensive, and complicated task in the ISA-18.2 Life Cycle. For existing systems, this step examines existing alarms to ensure they meet the intent and practices of the Alarm Philosophy Document. The result is generally a massive list of needed changes.

From hundreds of successful alarm improvement projects, this presentation shows the surprising (but very common) real world situations we have found. Attendees will be able to uncover similar examples in their own plants, helping justify alarm improvement efforts.

The problems found include:
• Poor Management of Change, even among highly respected and capable companies
• System and Alarm Documentation is often poor
• Highly important alarms are missing on process graphics; incorrect alarms are on them
• Poorly updated P&IDs have large quantities of important errors
• Redundant instruments are configured with different ranges
• Important safety interlocks have no pre-alarms
• Undocumented logic implementations are in the DCS, as well as logics that do not match the loop drawings
• Inconsistent alarm prioritization, done for surprising CYA reasons
• Important alarms that are suppressed with “no one knowing about it”

Common quotes from the alarm system owners:
• “I can’t believe we didn’t have an alarm configured on that.”
• “Ok, I know how it’s drawn, but that’s not how it actually is.”
• “That hasn’t worked in years.”
• “That’s been bypassed for ages.”
• “Our P&IDs are a patchwork of long-since abandoned processes and improvements.”
• “Our DCS modification philosophy has been to never remove a tag, just make another one.”