A Deming-Goldratt take on Risk Management
If we could predict the outcome of everything we do with mathematical certainty, then our lives would be much simpler. In reality, we exist within a complex web of interdependencies and such certainty is not possible. Hence, we live with uncertainty and its inevitable companion: risk.
We perceive a risk any time we feel that the path towards the objectives we are pursuing is fraught with uncertainty. If we felt no uncertainty, we would perceive no risk. This is concept of risk as defined, for example, by the ISO standards. It is closely connected with the concept of uncertainty, i.e. the inability to predict in a deterministic, Newtonian manner the outcome of an action or process.
Tackling uncertainty in a practical way
The monumental, and largely untapped or misunderstood, bodies of work of Dr. W. Edwards Deming (TPK) and Dr. Eliyahu Goldratt (TOC) deal explicitly with the issue of uncertainty in a paradigm shifting way: by focusing on how we measure facts and deal with perceptions. This provides a fundamentally practical approach.
Why is it so practical? Because from a Deming-Goldratt perspective it only makes sense to talk about “uncertainty” within the framework of process management, i.e. structured and intrinsically repeatable human activities, typically the ones that make up the life of an organization, NOT random events or possible scenarios.
In other words, a purely probabilistic approach to risk evaluation, while in some cases useful, does not provide any usable guideline to assess and manage systemic risks for organizations.
Management and prediction
For Dr. Deming any organization is a system, i.e. a network of interdependent components all aiming at one well understood and shared goal. No clarity on the goal, no system. Deming made it very clear that the key to understanding, measuring and managing a system is through the variation of its components, i.e. its processes.
The essence of management then, according to Deming, is the ability to “predict” the outcome of the processes making up the system. Prediction IS NOT forecast; it is rooted in statistical understanding and it entails the embracing of a “range of oscillation” (variation) as the outcome of a measure rather than a deterministic number (forecast). “Uncertainty” can then be defined as the range of oscillation of a statistically measured variable.
Managing risk in an organization means, essentially, managing that oscillation; the Theory of Variation, formulated by Dr. Walter Shewhart in the 1920s, helps to do precisely that through its major offspring, Statistical Process Control, SPC. (See our article on Variation) When we understand how processes are behaving through statistical knowledge then we know in advance what the outcome of a process will be.
In short: if we want to manage risk effectively in an organization we must abandon the idea of assigning a number (or a percentage) to it; we must, instead, understand the nature and the amount of oscillation (variation) the processes making up the organizational system are affected by and act according to the knowledge provided by SPC.
In our next post we will look at the role of a “constraint” in managing risk in an organization.
About the Authors
Angela Montgomery Ph.D. is Partner and Co-founder of Intelligent Management, founded by Dr. Domenico Lepore. Dr. Montgomery’s new business novel+ website The Human Constraint looks at how Deming and the Theory of Constraints can create the organization of the future, based on collaboration, network and social innovation.