From individual behaviour to system weaknesses: The re-design of the Just Culture process in an international energy company. A case study

This 2018 paper explored the effects of a re-designed Just Culture (JC) process in an international energy company.

It’s a solid read and worth your time – I’ve had to skip heaps.

Providing background:

·???????? This organisation started using the James Reason JC framework in 201, but collective experience in the organisation “indicated that the original design contributed to the misapplication of the JC process and in some cases led to outcomes not aligned with the intent”

·???????? The re-design took more influences from cognitive and social psychology, organisational justice and systems thinking

·???????? The re-designed JC process consisted of 8 questions to be used after an incident investigation

·???????? Results of this JC process were “to help shift the mind-set in the organisation and reinforce managers' understanding that most non-conformant behaviours are influenced by system weaknesses” and strengthening trust

·???????? In Reason’s version, a number of technical terms were employed that differentiated between types of violations and errors, supposedly linked with different levels of personal culpability

·???????? A diagonal line at the bottom of the process indicated diminishing culpability, “suggesting that behaviours on the left-hand side, e.g. sabotage, are more blameworthy than the behaviours on the right-hand side, e.g. system-induced or blameless error”

·???????? They argue that these decision flows could, however, “be used against the parties involved in an incident, if it is simplistically applied”

·???????? They give the example of how something deemed ‘possible reckless violation’ (not following rules) could be “portrayed as deserving a very similar level of blame as a “system induced violation” (rules not available or incorrect)”

·???????? Even terminology, like ‘violations’ has arguable value in these types of processes and in any case, a violation could be deliberate but not necessarily reprehensible, like departures from rules deemed necessary to maintain safe operations

·???????? “It is not surprising – although far from Reason's intent - that operators in some organisations coined a phrase “to be Just Cultured”, which implied “a punitive element to the process”

·???????? Hence, there is questionable fairness to the process as it is applied or perceived to apply in organisations

?

Discussing the ‘old’ JC process:

·???????? There appeared to be a focus solely on the person involved rather than prompting “deeper investigation into causes of the undesired behaviour and relied strongly on the use of punishment”

·???????? Providing an example of a gas release after a technician opened a valve, it’s said that the employee’s apparent misunderstanding of the situation “was interpreted by the management as “reckless violation” and led to a reprimand and not necessarily to a preventative corrective action”

·???????? In some areas of the business, the old JC process was implemented and used as designed, whereas in other areas it was associated with discipline and discredited

?

They discuss how the new JC process works – again I’ve skipped heaps. Some points though:

1.?????? The site leader gets approval from area manager to perform the JC review

2.?????? Site leader reads the incident investigation and seeks answers to the JC questions

3.?????? If the investigation report doesn’t have the necessary information to allow the JC questions to be answered then the leader goes back to the investigation team for the info

4.?????? At this stage, HR may be consulted on whether disciplinary measures are appropriate

5.?????? And several more steps that I’ve omitted.

New process:

Evaluation of the “new” Just Culture process

After evaluating 353 cases of the re-designed JC process:

·???????? 30% were related to the question “Did they understand what was required, and did they have the knowledge, experience, skill, physical capacity and resource to do it?” Examples included multitasking and an individual doing the job for the first time”

·???????? 24% were attributed to the question “Was the expectation clear?”

·???????? 21% were attributed to the question “Did they intend to act in line with expectations, but made a mistake?” Examples included mistakes provoked by plant configuration and multiple alarms

·???????? 10% related to “Is there evidence to suggest they acted to help self, company, to save time or effort?” Examples included work without a permit and crossing a barrier

They found that the corrective actions resulting from the JC shifted after implementation of the re-designed process. While the previous iteration “emphasised training, coaching or disciplinary actions”, the new process “emphasises addressing system defects, e.g. poor quality of procedures or equipment”.

55% of these actions related to system improvements, like updating procedures, enabling clearer expectations by line management, amending ?planning process or fixing equipment.

For the questions 1-6 which referred to system questions, coaching accounted for 35% of the actions, namely focused around “Did they understand what was required”, and “did they have the knowledge, experience, skill, physical capacity and resource to do it?”.

The coaching actions often tended to factors like people speaking up about resource or competency issues.

The level of coaching actions resulting from the JC were “higher than expected given that system issues account for 90% of the primary underlying contributing factors”. However, coaching actions often included things like the encouragement to speak up about system issues, like workload, lack of experience or other.

They also found via feedback that there was an ingoing confusion between the boundaries of the JC process and incident investigation processes. Hence, they created “official guidance on the boundaries” – see below:

Reflecting on the impact of the re-design process, they suggest that “it has supported an evolving shift in how leaders responded to incidents”.

Specifically, it helped to provide a clearer understanding of how context and situational factors affect performance and behaviour, and that: “the majority of human actions which contributed to incidents were due to various underlying system issues or weaknesses”.

They also report that following the re-designed process, leaders more consistently expected that the incident investigations explore the context of human performance and decisions, and understand the contributory factors.

Resultingly, the business also developed and communicated a set of human performance principles, as below:

Summing up some key points, they note:

·???????? The original JC framework was predominately based on UK criminal law concepts; which these authors challenged “whether this legal terminology should be uncritically applied in organisations, which, unlike societies, have control over selection, recruitment, supervision, verification and a plethora of other tools and processes”

·???????? The re-designed JC process doesn’t use the word violation, since this can be “easily misinterpreted and potentially lead to incorrect judgement of culpability”

·???????? Its in alignment with the philosophy that human operators, “although most proximate to an incident, is only one of many roles whose decisions may contribute to a situation leading to an incident” and hence it’s important to explore the other influences, including from leaders

·???????? The use of the new process is restricted to only those with training

·???????? A database has been implemented to allow learning from the JC reviews

·???????? Establishing clear boundaries between the JC process and incident investigations was important, to prevent the JC process being used as an incident investigational tool

·???????? Their findings are “a powerful example of the “What-You-Look- For-Is-What-You-Find” principle (Lundberg et al., 2009) demonstrating that the assumption about causes of accidents - in this case causes of behaviour - inform the questions asked and as a result, determine the investigation findings”

·???????? Hence, because the old process emphasised the role of negligence and recklessness in procedural non-conformance, the findings from the reviews necessarily “found” instances of this apparent recklessness

·???????? In contrast, the new process emphases the role of situational factors which shape actions and decisions over time

Link in comments.

Authors: Bitar, F. K., Chadwick-Jones, D., Nazaruk, M., & Boodhai, C. (2018). From individual behaviour to system weaknesses: The re-design of the Just Culture process in an international energy company. A case study.?Journal of Loss Prevention in the Process Industries,?55, 267-282.