Articles on AI, project risk management, probabilistic thinking, and decision making under uncertainty.
For the events that matter most - a spill, a rupture, a blowout - you rarely own enough history to estimate how often they happen. So you borrow the frequency from someone else's data, with the Poisson distribution and a real oil-spill dataset as the worked case.
Read article →
Most practitioners reach for a triangular or PERT distribution by default. Here is why that is structurally wrong for scenario-driven risk problems - and what to use instead.
Read article →
Companies that anchored their annual plans to a tariff rate are now managing a distribution they never modelled. This is not a forecasting failure - it is a category error. Why exporters and project finance professionals need dynamic probabilistic models, not updated spreadsheets.
Read article →
Reservoir engineers, wind-yield consultants, tunnelling contractors, actuaries - experts across industries hand you percentiles rather than distribution parameters. Here is the @RISK function that respects that, with a worked offshore wind example you can translate to your own sector.
Read article →
Sometimes the most honest thing the model can say about a variable is that it sits between two bounds and you have no idea where. Here is the @RISK function that encodes exactly that, without inventing information.
Read article →
The Strait of Hormuz is closed. Qatar has declared Force Majeure on 20% of global LNG. Forecasts are everywhere. Decision frameworks are not. This article builds one - combining a scenario probability tree, a BMMRJD time series model in @RISK, and Bayesian updating so you know exactly when to change your mind.
Read article →
Look at what is happening in the Persian Gulf and tell me you still believe in the single-point forecast. Three scenarios, three radically different futures - and organisations that do not model probability distributions are not managing risk, they are ignoring it.
Read article →
Teams adopt AI and move faster. Then something goes wrong - a risk flagged but never truly understood, a cost estimate that looked rigorous but wasn't. The research is now explicit: outsource the thinking, and you lose the capacity to think. Here is what correct AI use looks like in project risk management.
Read article →
AI systems are crossing a threshold: from augmenting analysis to participating in coordination, interpretation, and decision authority. Drawing on Choudary's concept of "restacking," this article examines who holds agency when humans and AI work together on project decisions.
Read article →
AI agents can now access files, reorganise folders, update spreadsheets, and execute multi-step tasks with minimal supervision. This is no longer speculation - it is operational reality, and exactly where structured risk thinking must step in.
Read article →
ChatGPT is not the last word in AI, just as Google was not the last word in search. The real asset is not which tool you choose - it is the skill to work with AI safely, fluently, and repeatedly inside real workflows.
Read article →
As learning algorithms become embedded in cost forecasting and risk prioritisation, they subtly alter authority, accountability, and professional judgment. The key challenge is no longer technical accuracy - it is how to work responsibly with systems whose logic cannot be fully inspected.
Read article →
Project management has no shortage of frameworks - yet outcomes remain inconsistent. The missing link is not methodology, but mindset. When probabilistic methods, simulation, and AI are integrated into everyday decision-making, the result is not certainty but resilience.
Read article →