Quantitative risk virtual summit 3rd edition

Details

14:00 🎤 Welcome Remarks – Craig Ferri, Timour Koupeev - 15 min

14:15 📊 Francois Joubert, "Using an “uncertainty matrix” to determine where risk sits in a project"

Abstract: It may happen that tornado graphs cannot provide useful information about a model's uncertainty. This may be caused by risk factors which may apply similar levels of uncertainty to the various line items in a bill of quantities. This causes several problems for project management professionals, as the question of “What matters most?” and “What do we need to focus on” cannot be answered by using a tornado graph. This problem may be solved by combining existing ModelRisk functionality and a cost uncertainty matrix to determine cost contingency per project package. The outcome of provides some insight into where cost uncertainty on the project.

15:00 ⏸️ Pause – 5 min

15:05 🎯 Rafael Nassur, "Prioritizing Task Selection: a new AHP Gaussian approach"

Abstract: This talk explores how Gaussian-based AHP and Monte Carlo simulation can be combined to transform Quantitative Risk Analysis results into a structured, probabilistic decision-making framework. By propagating uncertainty through both preference weighting and project outcomes, the approach bridges qualitative judgment and quantitative risk, enabling decisions that are transparent, consistent, and robust under uncertainty.

16:00 ⏸️ Pause – 5 min

16:05 🏭 Peter Vanneck, "Embedding Decision Quality in Strategic Business Decisions with ModelRisk"

Abstract: Strategic decisions are often weakened by cognitive and emotional biases that distort uncertainty and create false confidence. This session shows how biases like overconfidence, anchoring, and optimism systematically undermine decision quality, and introduces the Decision Quality framework to address this gap. We argue that probabilistic modelling with tools like Vose’s ModelRisk is essential—not optional—for rigorous strategic decisions, demonstrating how it exposes bias, reveals true value drivers, and turns intuition into transparent, defensible choices management can trust.

16:45 ⏸️ Pause – 5 min

17:50 🔹 Guz Vinueza, "Exploring Monte Carlo and Machine Learning"

An interesting take on how Monte Carlo simulation and machine learning can work together as complementary tools, where simulation provides probabilistic structure and ML extracts hidden patterns from complex risk data. Through practical examples using CART and k-means, it shows how these techniques improve variable selection, reveal natural groupings, and support the generation of meaningful correlations for more insightful risk-based decisions.

👋 Closing Remarks – Craig Ferri