Originally posted on January 16, 2015 @ 11:36 AM

Balancing Tight and Loose Coupled Systems

My last two articles on art and helping, generated feedback and questions that are worthy of further explanation. Amongst the feedback was that I ‘systems bash’ when I write about risk and safety.

Firstly, while I can understand how people may feel that I don’t value systems, however, the short response is, that I do. ‘Systems’ are helpful and necessary, they can create standardised actions and processes that effectively allow us to operate on ‘autopilot’ to help us get through our everyday life and live in society. For example:

· Driving (in Australia) on the left hand side of the road

· Lines painted in car parks so that we can park in an orderly fashion

· Forming lines at the supermarket before we pay

While I value (and need!) systems, I’m wondering if we need to consider them in a different way to that which is normally considered in risk and safety? To start this line of thinking, I thought it would be a useful to share my current thesis (worldview).

Foundational to my view is that all organisations and people, have to deal with, and make sense of risk, equivocality and subjectivity, despite a desire for certainty, clarity and objectivity. This is at the heart of the challenge for those of us in risk and safety. We desperately seek clarity and objective answers, views and process, yet the world is full of grey, messiness and bias.

This is why we are so focused on systematic approaches in risk and safety, we want to eliminate the grey and create standardization and control. Could it be that in order for systems to be more effective though, that we need to find a balance between regulation, adaptability and freedom so that people feel more in control of what they do, and hence will be more motivated?

Karl Weick in his book Managing the Unexpected introduces the concept of ‘loosely and ‘tightly coupled’ systems:

Coupling concerns the degree to which actions in one part of the system directly and immediately affect other parts. A loosely couple system is one where delays may occur and alternative pathways to completion are possible. A tightly couple system has little slack, and a process or set of activities, once initiated, proceeds rapidly and irreversibly to know or unknown conclusion. (2007, p.91)

The desire for tightly coupled systems can create angst for those who seek clear and consistent ways to do things (i.e. tightly coupled systems). Additionally, it creates real problems when we seek to control others, as this is demotivating, dehumanizing and can lead to the situation where people are not required to think for themselves, when a standardized process is the only thing accepted. So, perhaps the key question is, do we in risk and safety need to ‘let go a little’, and balance out what are mostly tightly coupled systems and create opportunities for ‘looseness’ so we are able to adapt to change as different circumstances arise? Will this help us be better prepared for the unexpected?

I hear you say, yes great theory and thanks, but how does that work in real life? What can we do to create balance between loose and tightly coupled systems?

Here are a few ideas that may assist:

· Engage in conversations, not just audit – Instead of a process that if focused solely on ‘auditing’ such as in a tightly coupled system, consider introducing a complimentary ‘loose’ system which focuses on engaging with people at the coalface (Weick refers to this as ‘sensitivity to operations’). Do this without an agenda or process and be aware of slipping back into a tightly coupled system which requires a standardised process – this is where I see Safety Conversations or Observations turn into KPI’s.

· Allow the right people to do the job – be open to change where required when choosing the people to do a particular job. In tightly coupled systems, responsibilities are very clearly stated and enacted. Where the system is ‘loose’, the decision about who does a task is not chosen from a ‘Responsibilities Matrix’, it is decided when needed and is based on the person with the best skills. Weick refers to this as ‘deference to expertise’.

· Allow ‘workarounds’ where necessary – while a process may require 15 steps to be carried out, it’s important to have, as Weick puts it, “a combination of keeping errors small and improvising workarounds that allow the system to keep functioning”. Of course the key to this is good relationships between those leading and those following so that these workarounds can be discussed and agreed when necessary.

I hope that this piece has helped those who feel that by adopting an approach that aims to better understand people and how we make decisions and judgments does not dismiss the role of systems.

As always, I look forward to the feedback.