A frequently reported strategy during alarm floods on distributed control systems is to have someone (another operator, engineer, plant manager) simply silence the onslaught of incoming alarms. If possible, the person silencing the alarm identifies an “important” alarm. The value of this strategy and the need for better design of plant alarms is documented in recent research from Ohio State University (Ho, C., Nikolic, M., Waters, M., and Sarter, N. “Not Now! Supporting Interruption Management by Indicating the Modality and Urgency of Pending Tasks.” Human Factors, Vol 46, #3, 2004).
The authors were investigating the problem of a new task interrupting air traffic controllers and the controllers’ need to manage their attentional resources. Research has shown that knowing the nature of a potential new task helps the controller decide whether their current task should be abandoned in favor of the new task. Additionally, information on the time required to complete the pending task further improved performance. The goal with the “interruption cue” is to prevent either dismissal of the pending task when it is more important than the current task, or attending to the new task when it is of less importance than the current task. The interruption cues (think alarms for process plants) work best if presented in a different modality than the task being performed (i.e., if the current task is visual, the cue should be auditory or tactile).
For process plants, since the primary task is largely visual (i.e., DCS displays), this new information about the competing task should be auditory. Isn’t this currently provided with alarm horns that match alarm priorities? In many cases, alarm horns have been disabled, because they’ve been deemed to be too much of a nuisance. However, at greater issue is whether the alarm priority actually conveys correct information on the importance of the new, competing situation. If the alarms do not actually reflect the seriousness of the pending situation and/or how quickly the operator must respond, then the information value is nil. Without someone (assuming they are available and knowledgeable) to read the descriptors as they roll across the screen, there is a good chance the competing task will either be ignored, or that it will interrupt a more important activity the operator is currently performing.
Reliance on an additional person being available to provide the information to the operator on pending alarms is a risky strategy. A better approach is to correctly prioritize the plant alarms based upon how fast the alarm must be responded to during off-normal events. With the availability of .wav files to generate auditory signals, even more information can be provided than with a set of tone generators. (Additional research is needed to determine how best to use this new capability.) Finally, it would be better if the interruption cue actually related to a task and not just myriad symptoms of the problem. For example, a “tower off-spec” or “tower upset” alarm is likely to be of more value than individual alarms on each temperature, level, flow, and pressure in the tower. However, this sort of higher level alarm also needs additional research in both its implementation and training requirements.
Copyright © 2005 Beville Engineering, Inc., All Rights Reserved
RELATED EXTERNAL MEDIA
|Consortium Reports New Findings on Alarm Rates||Automation World|
|How Many Alarms Can An Operator Handle||Chemical Processing|
|Impact of Alarm Rates and Interface Design on Operator Performance||Automation World|
|Operator Interfaces: Moving from Comfortable to Most Effective||Automation World|
|Operator Performance as a Function of Alarm Rate and Interface Design||Mesa.org|
This year's Fall meeting for the Center for Operator Performance will be October 24-26 in Corpus Christi. For more information, please contact Lisa Via. Guests are always welcome!
Our summer newsletter is now available. Click here!
Take our short survey on operator span of control. Click here (new window)
David Strobhar's book, "Human Factors in Process Plant Operation," is now available in both hardcover and Kindle e-book.
Copyright © 1996-2018 Beville Engineering, Inc. All rights reserved. (937)434-1093. Beville@Beville.com