ESSENTIALS OF SAFETY BLOG 14/14
There are a significant number of inputs to safety outcomes and many of them are able to be focussed upon and when needed, manipulated to increase the likelihood of things going well. Understanding the links and influences of these inputs greatly enhances our ability to understand and manage for safety in the workplace.
This is the fourteenth of a dozen or so blogs covering the Essentials of Safety that I talked about in the first blog of this series. We have covered an introduction – which we called Essentials of Safety, Understands their ‘Why’, Chooses and displays their attitude, Adopts a growth mindset – including a learning mindset, Has a high level of understanding and curiosity about how work is actually done, Understands their own and others’ expectations, Understands the Limitations and use of Situational Awareness, Listens Generously, Plans work using risk intelligence, Controls Risk, Applies a Non-Directive Coaching Style to Interactions, Has a resilient performance approach to systems development and Adopts an Authentic Leadership Approach When Leading Others.
The Oscillations of Safety in modern, complex workplaces – The Safety Oscillations Model
A model that greatly helped me understand and tweak a model ‘safety’ system was built after a review of Resilience Engineering: Concepts and Precepts, Edited by Erik Hollnagel, David D. Woods, and Nancy Leveson. CRC Press 2006. Specifically, it was a figure describing a simplified model of the dynamic behind the space shuttle Columbia loss. It is also used in a paper by Leveson in greater detail. I have generalised it to talk about the creation of safe work. I adapted it and it is used both to understand how the system elements influence other elements and how it can be used by safety professionals and business leaders to keep an eye on things and help them emphasise or de-emphasise activities for the purposes of getting it right. I wrote a simple paper on it and it is reproduced here:
We tend to measure safety outcomes as a surrogate for safety. In the model, I have called a difference between ‘safe outcome’ as an outcome and ‘in-process safety’ – stuff that makes things go right on a day-to-day basis (Figure 1).
There are a number of inputs to safety outcomes and in-process safety. Some are closely coupled and some more distant. Some of the more closely coupled include: effort on safety, production pressure, budget for safety, and priorities for safety. Some of the more distant couplings include: business budget, external safety focus, external production drivers and risk view. I have attempted to capture them in the flow diagram above, which highlights the buttons that can be pushed to drive things in the direction we want and away from things we do not want.
Note: The darker lines represent something that we could focus on – maximise/strengthen, and the dashed lines represent things that we could minimise – push back against. The simple solid lines represent simply how one thing will/can lead to another.
So, let’s start with the ‘Safe Outcomes’. If ‘Safe Outcomes’ goes up (‘Safety’ gets better such as a decrease in injuries), represented by a + sign on the right of the ‘Safe Outcomes’ box, then the laps start:
An increase in ‘Safe Outcomes’ can lead to three different things. One is ‘Happy People’ – something we should maximise. Another is an increase in ‘Complacency’ as everything is going well. This is clearly a link we want to de-emphasise as much as we can. The third link is from an increase in ‘Safe Outcomes’ to an increase in ‘Production Expectations’ as ‘Safety’ is humming along nicely. This is another thing we want to minimise and de-emphasise.
Firstly let’s follow the increase in ‘Complacency’ link. An increase in ‘Complacency’ can lead to a reduction in ‘Effort for Safety’ – which leads to a decrease in ‘In-Process Safety’ and then an increase in ‘Risk’, an increase in ‘Incident Rate’ and hence a reduction in ‘Safe Outcomes’. Continuing the right-hand loop for now, a decrease in ‘Safe Outcomes’ can lead to a decrease in ‘Complacency’ – which we want to encourage. This can result, with some help/pushing, in an increase in ‘Effort for Safety’ and hence an increase in ‘In-Process Safety’, a subsequent lowering of ‘Risk’, a reduction in ‘Incident Rate’ and hence an increase in ‘Safe Outcomes’. Which is where it all started. This oscillation is never ending but also has other inputs and impacts.
Another previously mentioned result from an increase in ‘Safe Outcomes’ is an increase in ‘Production Expectations’. This needs to be minimised as we talked about earlier. An increase in ‘Production Expectations’ leads to an increase in ‘Production Pressure’. This leads to a decrease in ‘Effort for Safety’ and around the right-hand loop we go again.
If we now look at one of the more distant couplings. ‘Business Budget’ for example. An increase in ‘Business Budget’ will result in, with the right emphasis and push, in an increase in ‘Budget for Safety’. This will result in an increase in the ‘Effort for Safety’ which results in an increase in ‘In-Process Safety’ and ‘Safe Outcome’. Around and around the mulberry bush we go again. You can work out the rest.
We can use a tool from resilience engineering to keep an eye on things. If we build our safety capability to monitor and detect the weak signals within each ‘element’ so that we can alert when the element gets close to a failure boundary and move the emphasis away from the boundary. We can then direct effort to either emphasise or de-emphasise the link to the next element and hence impact ‘In-Process Safety’ and ‘Safe Outcome’ directly and/or indirectly by maximising and minimising the links as required’.
Key Takeaway: There are a significant number of inputs to safety outcomes and many of them are able to be focussed upon and when needed, manipulated to increase the likelihood of things going well. Understanding the links and influences of these inputs greatly enhances our ability to understand and manage for safety in the workplace.