The following slides are extracts from Cognitive-Edge’s Creating and Leading a Resilient Safety Culture course.
The first safety manual was written in 1933 during the Age of technology. Next came there recognition of humans, safety management and systems. The latest age appreciates how working safely in today’s VUCA environment requires an understanding of complex adaptive systems.
A company’s safety program will typically include elements of all 4 ages. The pieces define the culture and how “things get done safely around here.”
The S-curve graphic below is adapted from a presentation made by Erik Hollnagel at a May 2012 Resilience Learning Lab in Vancouver and a tutorial given to the Resilience Healthcare Learning Network in June 2012. The first 3 ages are essentially grounded in the Theory of Error.
Things can go wrong because technology fails. Or humans fail. Or organizations fail. Safety-I requires the ability to prevent something from going wrong.
This is achieved by:
1. Finding the causes of what goes wrong (Root Cause Analysis).
2. Eliminating causes, disable possible cause-effect links.
3. Measuring results by the # of fewer things going wrong.
The Safety-II paradigm asks the question: Why only look at what goes wrong?
Safety-II is defined by the ability to succeed under varying conditions. Humans are seen as heroes creating safety and not cogs in some socio-technical system.
Adjustments by humans are the reason why everyday work is safe and effective. It’s also the reason why things sometimes go bad despite the efforts of humans. Therefore, in Safety-II, the measures of success are on impacts and not on outcomes.
The Cynefin Framework validates many prevailing safety practices as useful and effective when appropriately applied.
If a safety practice is Cause & Effect focused, it resides on the Ordered side. Tools such as Safety Management Systems and Bowtie risk management which require a Safety professional to manage are in the Complicated domain.
The Domino Theory model was the first scientific approach to accident prevention. The cause & effect model reduces the problem into distinct parts. James Reason’s Swiss Cheese model is another reductionist approach. The popular Hearts & Minds safety concept depicts safety evolving in a linear fashion. The trouble arises when these models are applied to the Cynefin Unordered side where cause & effect relationships are either unknown, unknowable or non-existing due to workplace volatility, uncertainty, confusion, unpredictability, ambiguity.
Robustness is taking a Safety-I approach to prevent something from going wrong in the Complicated and Simple domains.
Resilience requires a Safety-II approach because a work-as-imagined socio-technical system is unable to deal with unknowable and unimaginable situations. Only humans can and will vary performance accordingly.
Operationalizing the Safety-II paradigm has been a struggle. The primary reason as explained in the Berkana Institute’s Two Loops video is because the current paradigm (i.e., Safety-I) is so dominating.
In the Age of cognitive complexity, there is a new view where Safety is an emergent property of a complex adaptive system. Erik Hollnagel states:
“We must face the fact that the world cannot be explained by cause-effect models. Incidents and accidents do not only happen in a linear way, but include emergent phenomena stemming from the complexity of the overall system.”
This view of safety helps to explain why rules are two-edged. When the working conditions have just enough rules, safety emerges. However, when there are too many rules, danger emerges in the form of worker frustration leading to deliberately ignoring. When the tipping point is reached, failure occurs in the form of an accident.
Hollnagel E., Wears R.L. and Braithwaite J. From Safety-I to Safety-II: A White Paper. The Resilient Health Care Net: Published simultaneously by the University of Southern Denmark, University of Florida, USA, and Macquarie University, Australia. 2015.