This involved two Human Factors Safety Critical Task Analysis (SCTA) studies: 1) the first was on a process concerned with the inadvertent charging of drummed materials, 2) the second was on a process concerned with the inadvertent charging of bags of powdered materials. Essentially, mixing materials that could lead to a major event.
The small batch processing plant had identified two processes through their ongoing risk assessments that contained risks that relied on human actions. If these actions were performed incorrectly they could result in Major Accident Hazard outcomes. The risk assessment included assumptions about the controls and PIFs that reduced the risk of these events occurring. Given the high consequences and the vulnerability to human failure they decided further Human Factors scrutiny was required.
The first study involved the charging of drummed materials. The existing risk assessment had assessed the likelihood of inadvertently charging the incorrect substance as low. This was because: a) the dangerous substance was never required in this room, and b) there were physical connections specific to the correct substance.
The second study involved the charging of powdered materials. This was quite different as four reactors, which all looked similar and were in close proximity, were charged with four different powders that were all bagged and stored together in a space next to the reactors. So material not to be charged was in the vicinity and could be confused. The existing risk assessment took credit for the materials being delivered in different sized bags (making them harder to confuse).
The small batch processing plant had identified two processes through their ongoing risk assessments that contained risks that relied on human actions. If these actions were performed incorrectly they could result in Major Accident Hazard outcomes. The risk assessment included assumptions about the controls and PIFs that reduced the risk of these events occurring. Given the high consequences and the vulnerability to human failure they decided further Human Factors scrutiny was required.
The first study involved the charging of drummed materials. The existing risk assessment had assessed the likelihood of inadvertently charging the incorrect substance as low. This was because: a) the dangerous substance was never required in this room, and b) there were physical connections specific to the correct substance.
The second study involved the charging of powdered materials. This was quite different as four reactors, which all looked similar and were in close proximity, were charged with four different powders that were all bagged and stored together in a space next to the reactors. So material not to be charged was in the vicinity and could be confused. The existing risk assessment took credit for the materials being delivered in different sized bags (making them harder to confuse).
For the first study we focused on the Human Factors issues that could have led to confusing the drums. We performed a walk-through talk-through of the process from drum delivery to site, through storage, selection and then charging. This was relatively well controlled although some potential improvements in checks and signage were identified here.
When examining the charging process, we found that it was relatively easy to inadvertently bypass the specific physical connections (which the risk assessment had taken credit for) as there were different ways in which hoses could be connected. Issues to do with the robustness of checking process and emergency response were also explored.
For the second study we did not go as far back in the process because the bags were co-located before charging anyway. We also found that the bags were not different sizes as expected by the risk assessment. This increased the likelihood of error.
However, we also found a strong positive PIF that was not discussed in the risk assessment. The four reactors that were next to each other A, B, C and D, must be charged with the corresponding bags of powder A, B, C and D. The dangerous scenario was putting bag B in reactor A. We found that two different departments worked on these reactors, i.e. department 1 worked on reactor A exclusively, while department 2 worked on reactors B, C and D. In addition, material A was usually charged in a liquid rather than powder form. Both of these factors decreased the likelihood of error.
Further recommendations were made to increase the separation of bag A from B, C and D, to differentiate the reactors, to introduce automatic detection of the wrong material and to improve the manual testing of the material before it moved on further into the process (where the worst consequences would be realised).
For the first study we focused on the Human Factors issues that could have led to confusing the drums. We performed a walk-through talk-through of the process from drum delivery to site, through storage, selection and then charging. This was relatively well controlled although some potential improvements in checks and signage were identified here.
When examining the charging process, we found that it was relatively easy to inadvertently bypass the specific physical connections (which the risk assessment had taken credit for) as there were different ways in which hoses could be connected. Issues to do with the robustness of checking process and emergency response were also explored.
For the second study we did not go as far back in the process because the bags were co-located before charging anyway. We also found that the bags were not different sizes as expected by the risk assessment. This increased the likelihood of error.
However, we also found a strong positive PIF that was not discussed in the risk assessment. The four reactors that were next to each other A, B, C and D, must be charged with the corresponding bags of powder A, B, C and D. The dangerous scenario was putting bag B in reactor A. We found that two different departments worked on these reactors, i.e. department 1 worked on reactor A exclusively, while department 2 worked on reactors B, C and D. In addition, material A was usually charged in a liquid rather than powder form. Both of these factors decreased the likelihood of error.
Further recommendations were made to increase the separation of bag A from B, C and D, to differentiate the reactors, to introduce automatic detection of the wrong material and to improve the manual testing of the material before it moved on further into the process (where the worst consequences would be realised).
Get the latest newsletter by signing-up today
This free 30 minute mini-course will introduce you to Human Factors, and how critical task reviews are used to improve the quality and safety of tasks and processes across different industries.
It’s free, informative and you’ll even get a certificate of completion.
This short and engaging handbook provides a great overview of Human Factors Systems Critical Task Analysis (SCTA) and how it helps people across sectors reduce error and improve human performance.
SCTA can help keep people safe and delivers value.
Popular Pages
© Human Reliability 2023.
1 School House, Higher Lane,
Dalton, Lancashire,
WN8 7RP, UK.
T: +44 (0) 1257 463 121
E: [email protected]
