• Zachary Collier

The Role of Uncertainty in Decision Making

One of the biggest reasons why decision making is difficult is because of uncertainty. If we could peer into a crystal ball and see the future with certainty, decision making would be easy. But in reality, there are a lot of things that we just don’t know, and often it is impractical to gather all of the information necessary in order to achieve certainty. Many decisions therefore have to be made under uncertainty with the best information we have at the time.

What exactly is uncertainty? The Society for Risk Analysis proposes two related definitions (1):

- For a person or a group of persons, not knowing the true value of a quantity or the future consequences of an activity

- Imperfect or incomplete information/knowledge about a hypothesis, a quantity, or the occurrence of an event

Both definitions essentially point to the same thing, which is a lack of information and knowledge about something relevant to decision making. Generally speaking, there are two types of uncertainty – uncertainty due to the natural variability in the world, and uncertainty due to a lack of knowledge. (If you want to impress your friends at a fancy cocktail party, you could use the technical terms aleatory and epistemic uncertainty for variability and knowledge uncertainty, respectively).

Variability is related to the randomness that we find in the world, especially when we sample values from a population. Values that are subject to variability are uncertain regardless of how much additional information is gathered (2). Variabilities can arise due to fluctuations across time and space, as well as between heterogeneous items (e.g., different insects have different resistances to pesticides) (3).

Knowledge uncertainty, on the other hand, can in theory be reduced by acquiring more information. Not knowing who the Vice President of the United States was in the year 1900 is an example of knowledge uncertainty. This type of uncertainty can be reduced by researching the relevant sources, although in practice it might be difficult to find. Knowledge uncertainty comes up when we build models to solve problems (3). There may be uncertainty in the structure of the model itself – does it accurately correspond to the real-world system or problem? Other knowledge uncertainties arise in the parameter values of the factors of the model (e.g., what discount rate is the correct one to use).

Why does all of this matter?

Understanding uncertainty is important because decisions are almost always subject to at least some uncertainty, and knowing how to navigate through uncertainty can help your organization make better decisions. Importantly, being able to recognize different types of uncertainty can point to different uncertainty reduction strategies. For example, when dealing with variability, collecting a larger sample may or may not help reduce the uncertainty but will never fully eliminate it, whereas more research may be able to fully eliminate certain sources of knowledge uncertainty. Probabilistic modeling techniques can be used when there is variability, but if you don’t understand the problem being modeled (model uncertainty), then chances are that your model results won’t be very meaningful, regardless of how much data you gather.

Finally, understanding the uncertainties that are most relevant to your decision can aid in determining how much of your resources you should be willing to expend to reduce uncertainty. Additionally, uncertainty can be used to build flexibility and options into your decision making, using uncertainty to your advantage.

Collier Research Systems offers consulting services to help navigate through the uncertainty that clouds decision making. To learn more, visit


(1) Society for Risk Analysis.

(2) O’Hagan, T. 2004. Dicing with the unknown. Significance, 1(3), 132-133.

(3) Haimes, Y.Y. 2016. Risk Modeling, Assessment, and Management, 4th Edition. Wiley.