The Weekly Reflektion Week 49 / 2019

This week’s Reflektion is inspired by a presentation on Causal Learning that we recently attended and some reflections around the concept of ‘human error’.

Do you have ‘human error’ down as one of the main causes for accidents and incidents?

Maybe you should think about the cause of human error rather than human error as a cause before assigning actions to prevent recurrence.

For the Reflektion on Three Mile Island in week 46 we talked about the ‘mindset’ that technical systems are designed satisfactorily and that any person involved in the operation would only improve the situation. We considered that this fails to recognize the fallibility and limitations of people and their capacity to do things that are not expected.

In the Reflektion on the Southhall train crash in week 48 we stated that systems should be designed and operated such that a single fault or failure should not lead to a serious incident or major accident. In the event that safety systems are disabled or inhibited, mitigation measures should be put in place to maintain this principle. We emphasized that current views on human error recognize the fallibility of the operator and the inevitability of mistakes.

In both of these Reflektions, we are stating that people can make mistakes, and implying the inevitability of ‘human error’. We are emphasizing that the possibility of ‘human error’ be considered in the design and operation of systems. We are not however considering the causes of ‘human error’.

Trevor Kletz, a safety advisor for the Institution of Chemical Engineers once said, concluding that an incident was caused by human error is the same as saying a dropped object is caused by gravity. It may be true but it is unhelpful in the understanding of why the incident occurred. Sidney Dekker in his book ‘The Field Guide to Understanding Human Error’ encourages us to see the identification of human error as a start point for finding the causes and not a cause in itself.

Experience shows that actions taken in response to incidents often do not lead to changes in performance that prevent the incident occurring again. One of the reasons is that actions intended to change behaviours of individuals are often implemented without a proper understanding of the causes of the behaviour.

The ‘Causal Learning’ methodology for investigations could contribute to a better understanding of why incidents happen and identification of measures to prevent recurrence. ‘Causal Learning’ combines two basic ideas. The first is an investigation process based on causal reasoning, shown in the outer circle of the above figure. The second is a guide to how humans learn, shown in the inner circle of the above figure. 

Traditional investigation techniques focus on what went wrong, what was missing or what should have been done. Traditional investigation techniques try to identify the errors and mistakes made by the people involved. This approach is driven by hindsight bias and leads to defensive reasoning. It does not really lead to an understanding of why an incident or accident occurred. Causal Learning has a focus on positive actions and actual conditions that were relevant to the incident. Causal learning starts with the premise that people follow their convictions when taking actions and do not intentionally do something wrong.

The measures identified from the Causal Learning investigations are based on how to improve the conditions and how to influence the convictions so that the individuals will make the right decisions and take the right actions next time.

Learning is the key to improvement in performance and prevention of incidents and accidents. Causal Learning defines learning as the creation of new understanding or belief that is different from a previous understanding of belief. Learning is therefore not just the acquisition of new knowledge but also the creation of new beliefs and convictions.  The process to create new beliefs is challenging and demanding for all concerned and in particular the management and supervisors.

One of the main reasons for not taking on learning challenges is demand on time. We are all very busy and the choice between sending out a safety flash and engaging in a learning process to create new beliefs seems pretty clear. The safety flash can be sent by email in a couple of minutes and the email itself is documentation that something has been done. The learning process on the other hand is a bit harder to predict and may lead to even more work and more time. However, consider how much time you spend on incident investigations, follow up of actions, reporting of incidents, reporting of action status, discussions with the senior management, the authorities etc etc etc? Perhaps an investment in Causal Learning will give you a positive return on your time.