Depth of Uncertainty in Project Management

Uncertainty in projects arises from unclear objectives, changing requirements, insufficient knowledge of project teams, external disruptions, and unpredictable risks, which often lead to delays, cost overruns, and inefficiency. In PMBOK, uncertainty is described as a lack of understanding and awareness of issues, events, path to follow, or solutions to pursue.

But deeper research identifies three categories of uncertainty that together cover all aspects of project management:

• Ontological (existential) uncertainty, which can be defined as a state of complete lack of understanding of the existence and purpose of the model of the relevant aspect of the system.
• Epistemic (knowledge) uncertainty, which is related to the lack of knowledge about the system model and the inaccurate encoding of the physical system in the model.
• Aleatory (randomness) uncertainty, which can be considered as the randomness of the process represented by the system model.

To effectively manage uncertainty during project decision-making, it is crucial to recognize the full spectrum of uncertainty levels, ranging from the ideal yet unattainable complete confidence and determinism to the opposite extreme of total ignorance and chaos. In the PMBOK Guide — 8th Edition, the range of levels of uncertainty and their challenge to decision makers ranges from “known knowns” to “unknown unknowns.” This framework is based on philosophical and practical discourses, from ancient Greek epistemology to modern decision theory, emphasizing the complexity and breadth of uncertainties.

In order to further study uncertainty and develop tools and methods for reducing uncertainty in projects, it is proposed to introduce a new unit of measurement for total uncertainty — 1 Trump. Each level denotes a progressive increase in uncertainty on a scale from 0.0 to 1.0 Trump:

• Complete confidence and determinism (0.0 Trump): This denotes a state where everything is known precisely, providing the basis of absolute determinism. This is an ideal practically impossible to achieve in real scenarios, but it serves as a limiting characteristic at one end of the spectrum.
• Uncertainty Level 1 (0.0–0.2 Trump): This level acknowledges minor uncertainty but does not require detailed measurement. These situations typically involve short-term decisions where there is sufficient historical data to predict outcomes. They represent simple “known unknowns” and are close to complete confidence on the Trump scale.
• Uncertainty Level 2 (0.2–0.4 Trump): Here, systems and input data can be assessed probabilistically, or future scenarios can be determined with sufficient accuracy and corresponding probabilities. This level includes “known unknowns,” where risks can be quantified using probabilities, and risk management methods can be used for decision-making.
• Uncertainty Level 3 (0.4–0.6 Trump): At this stage, although numerous probable future cases are recognized, exact probabilities cannot be assigned. Decisions are made using scenario analysis, exploring various possible future worlds without definitive probability, indicating increased uncertainty and, thus, a higher position on the Trump scale.
• Uncertainty Level 4: Divided into 4a and 4b, this level captures deep uncertainty:
  • 4a (0.6–0.8 Trump): Many probable future events can be outlined, but the exact models and probabilities of these futures are unknown due to limited data or understanding of the mechanics.
  • 4b (0.8–1.0 Trump): We only know that we do not know—this relates to unpredictable events, also known as “black swans,” which cannot be predicted by analyzing past data and are only recognized retrospectively.
• Total ignorance and chaos (1.0 Trump): Representing the opposite end of the spectrum from complete confidence, this level denotes a state of complete unawareness of future possibilities or impacts, constituting total randomness, unpredictability, and chaos, where participants have no way of knowing the full extent of their ignorance.

For more details, see the article “Categories of Uncertainty Affecting Project Management Information Systems.”