The report identifies human factors as a key part of process safety. It also raises a number of human factors issues, which are summarised below.
For most of its incident investigations, BP uses a list of causal factors to analyze root causes. BP refers to this method as the Comprehensive List of Causes (CLC).
A list of human factors is also provided for use in conjunction with the CLC. This contains a guide to analyzing human behaviors, beginning with a determination of whether the identified behavior leading to a cause was intentional or unintentional and leading to the identification of external and internal influences and other conditions under which personnel are likely to make mistakes.
In the Panel’s experience, investigations typically use a checklist as a complete list of potential causes instead of a starting point for discussion of the deeper root causes and usually will not identify factors that are not on the list.
The Panel also believes that BP’s list of systemic factors related to engineering problems (e.g., “inadequate technical design”) appears somewhat superficial.
While inadequate technical design is a valid factor, BP should use it to invite more extensive inquiry: What is the design inadequacy? Why was it present? Why was it not discovered prior to the incident under investigation?
Many of the listed systemic factors do not represent systemic issues. Fatigue, for instance, is included as a systemic cause.
BP uses the CLC for both personal safety accidents and process safety accidents. In the Panel’s opinion, the causal factors involved in occupational or personal safety incidents and process safety incidents typically are very different.
The human error analysis, which focuses investigators’ efforts on personal safety aspects of incidents rather than all aspects of an incident, may introduce additional bias in the analysis toward finding behavioral root causes.
At the time of the Carson refinery technical review in May 2006, about half of process hazard analysis, or PHA, action items at Carson from 2001-2004 remained open.
Action items from facility siting and human factors checklists used in PHAs were not consistently tracked and implemented.