Since 1975 a number of oil refineries have been shutdown in the US and no new refineries have been built. Yet, according to the Energy Information Administration, total petroleum consumption has increased 13%. To keep up with the demand, the remaining plants have been debottlenecked and added process units to increase capacity. Changes at the plants happened slowly and were incremental. Although each incremental change appeared to be insignificant when viewed in isolation, the sum of all of the changes has been substantial.
Beville has recently completed several studies in which plants had undergone considerable incremental equipment changes over the last twenty five years while the organizational structures and operating practices remained static. Operators at the plants complained that their plants had become too large, there was simply too much equipment to know and operate, and the sum of the changes were overwhelming. The personnel argued that the net result of all of the changes was a reduced level of safety in their plants.
Looking at all of the changes, the arguments seemed to be logical. Humans do have limitations in physical strength, stamina and cognitive ability. But was this the case; did the job expectations exceed what can reasonably be expected of operators? Are the operators “overloaded”, either during steady state or upset operation? Managers simply did not have enough data points for comparison to make valid assessments. As a result, Beville Engineering was asked to benchmark the plants to others and assess the validity of the safety concerns.
To investigate the operators’ complaints, Beville used a number of different techniques to benchmark the plants’ characteristics to others in the industry. First, a job sampling technique, similar to a time and motion study, was used to evaluate steady state workload. The workload benchmarking found that the workload parameters; including the percentage of time operators spent completing job activities, crew communications, and control system usage, were reasonable, and in some cases, well below industry averages.
Next, upset response workload was assessed through an upset response task analysis technique. For the assessment, operators talk/walk through the actions they take during a reasonable, worst case upset scenario. The loadings for each individual job were summed and divided by a target to derive the number of operators needed to respond for the plants. The upset response loadings were reasonable, but there were a number of high effort bottleneck tasks that hindered the crews’ upset response performance, some the result of pushing the plants past original design rates.
While the operators were interacting with their equipment during the steady state and upset assessments, the equipment operability was also being assessed. Equipment that was outside of optimal reach zones and valves that required high physical effort were identified. Equipment performance shaping factors, such as labeling and lighting, were also evaluated. Plant equipment integrity (such as grounding of equipment and motor shaft coupling guards) were assessed as were housekeeping/cleanliness and safety adherence. Deficiencies in equipment integrity indicate lapses in plant management, supervision, and operator motivation.
Another comparison evaluated the total amount of equipment under the operators’ control. Operators felt that they had more equipment responsibilities than other plants and that the total amount of equipment exceeded what could be expected of them to operate. The equipment comparisons found that the amount of equipment per position was average and significantly below those that were the highest.
Based upon the benchmarking comparisons, the concern that the amount of equipment in the plants exceeded the operator’s cognitive limitations did not appear to be valid. Across the board, the benchmarking measures were below industry averages. However, Beville associates did observe skill/knowledge deficiencies across the crews.
Upon review of the information collected, there appeared to be a number of causal factors including:
• The training system and training material were deficient.
• There were lapses in operational consistency and practices, such as crews choosing different courses of action in response to the same event.
• Foremen had become tied to the control board and interacted less with the outside operators than others in the industry.
• Adding to the problems were other human performance shaping factors, including equipment that was difficult to use and operate.
• The control system displays and alarms were also deficient and did not make use of software upgrades.
• While debottlenecking had increased plant capacity, it had also made securing the plants much more difficult.
It appeared that the summation of these and other factors were producing significant difficulties for the operators and giving them the sense that their jobs were overwhelming.
In the preceding examples, small incremental changes over the span of twenty years had produced significant changes in the plants’ operating characteristics. It is often difficult to see the results of the changes until their cumulative effect crosses a threshold and creates disgruntled operating crews. These accumulations of incremental changes require managers to periodically assess and rethink their plants operations and human factors. An excellent way to assess the changes is through comprehensive plant audits that benchmark the plants’ operating characteristics to others in the industry. In the preceding examples, the audits identified the deficiencies in human factors and changed the nature of the debate on how to improve the plants’ performance.
Copyright © 1998 Beville Engineering, Inc., All Rights Reserved
RELATED EXTERNAL MEDIA
|Control Systems, Oh, No...Not Again!||Control Global|
|David Strobhar: Breaking Down Human Language||Intellirights.com|
|DCS Console Operator Issues in Related Industries||2011 TAPPI PEERS Conference|
|How Good is Your Operator's Mental Model?||Mynah|
|How to Build a Better Operator - ABB Automation & Power World||Control Design.com|
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