True or false, high control board operator loadings are usually caused by unplanned abnormal events, such as equipment failures? True you say? Guess again. In talking about abnormal situation management, speculation has run rampant on the exact nature and frequency of unplanned abnormal events. What the debate is sorely lacking is some objective data that is sampled from across the industry from which sound decisions can be made. The reality, based on data from Beville Engineering’s job sampling database, is that high control board loadings are seldom caused by unplanned abnormal events. Furthermore, they are not necessarily always bad and, in fact, many high loadings are a reflection of operators coping with long stretches of low process activity and periods of boredom.
Beville Engineering, Inc. has been studying control board operator loading for the past two decades and compiled a database of job loadings from across the petrochemical industry. The data, collected through a variation of time and motion study that Beville refers to as job sampling, contains a collection of job related activities that operators were observed performing over the course of four hour sample periods. In the database, the job related activities are delineated into standardized task categories containing both the length of time for each task and its frequency. Each individual task category is summed to form an overall loading representing the total percentage of time spent completing job related tasks.
Currently, Beville’s database contains 225 control board operator samples. The samples were collected between 7-11 AM and 3-7 PM. The sampling periods are intended to capture the operators at their busiest, when maintenance personnel are beginning their activities, and a somewhat slower activity period in the afternoon when the operators are catching up on daily tasks. Graveyards are not sampled since the high end of the workload spectrum is the area of most concern.
Recently, the database was analyzed and the top quartile studied to find commonalties and trends in the causes of high job loading. The data were found to generally fall into eleven different causal categories. The following figure contains the data rank-ordered by number of samples falling into each of the categories. Of the fifty-seven samples from the top quartile, the most frequent cause of high loading, at fifteen occurrences, was due to high administrative/logging activities. The second highest cause, accounting for nine of the samples, was due to the operators completing special projects (material balance, revising drawings, etc.). Third highest, accounting for seven of the samples, was engineers running equipment tests (those pesky engineers, this is actually how Chernobyl started).
Surprisingly, unplanned equipment failures and upstream /downstream unit problems accounted for only 6 of the samples. In comparison to all of the control board samples, 4% of the samples could be classified as unplanned abnormal situations that significantly affected the board loadings over a four hour sample of time. Many of the events that could be classified as abnormal do not significantly increase the average loading over a four hour sample. One reason is that the positions are so under loaded that even with an abnormal event, the positions are still only marginally loaded. Other causes are that the operators are able to control the situation in a relatively short period of time, thus avoiding significant prolonged upsets.
In summary, most of the high control board operator loadings are the result of planned activities. Only a small percentage (4%) were the result of what could be called unplanned abnormal events (i.e., equipment failures or feed/rundown disturbances). However, small disturbances can end up costing millions of dollars in lost production or equipment damage.
The next logical step in the analysis is to examine how much the upsets added to the positions’ loadings. It is Beville’s practice to re-sample any position that has unusual activities occurring during the sampling. Our next newsletter will examine the difference between the upset samples and resamples and the implications for controller loading targets. Stay Tuned.
Copyright © 1997 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 © 1996-2019 Beville Engineering, Inc. All rights reserved. (937)434-1093. Beville@Beville.com