Thanks, Pegah.
Original Message:
Sent: 10-20-2025 11:08 AM
From: Pegah Farshadmanesh
Subject: Risk Modeling and Management for Natural Disasters: Innovation by the Numbers Presentation
Hi Mitch,
Thanks for the interesting points and questions!
2. You're absolutely right, risk evolves over time. Physical degradation and changing operating conditions can alter both likelihood and consequences. In harsh environments, materials and components are exposed to corrosion, fatigue, or thermal stresses that accelerate degradation. These effects not only increase the probability of failure but also elevate maintenance and inspection costs as systems age. Currently, integrating simulation tools that can model these degradation mechanisms enables more risk-informed decision-making by allowing engineers to anticipate and prioritize where failure is most likely to occur. However, from a regulatory standpoint, it is essential to account for all sources of uncertainty, including model assumptions, to ensure that decisions remain defensible and consistent with safety objectives over the system's lifecycle.
3. Maybe a large-scale safety-critical infrastructure, where multiple subsystems interact across physical and digital domains . The interdependence between structural, control, and operational systems is highly non-intuitive. An example can be the nuclear reactor safety. The interdependence between thermal-hydraulic, structural, and control systems is highly non-intuitive. We need to use PRA and system-level fault trees to capture failures that individual can't foresee. We need to consider human factors which add additional complexity to our analysis. Another example can be power grids, probably.
6. The Texas Hill Country flood tragedy, I believe, reflects risk normalization. I think repeated misses foster false confidence.
As a separate note here: I came across this interesting paper regarding overreliance of human on AI system that you maybe interested of (https://arxiv.org/abs/2212.06823) . 7. I think the level of assessment should depend on the potential consequences and external factors such as resource limitations. In some cases, it may be more practical to adopt a hybrid assessment approach that balances analytical rigor with available time and data (Also, level of expertise plays an important role in these scenarios).
Original Message:
Sent: 10/19/2025 5:37:00 PM
From: Mitchell Winkler
Subject: RE: Risk Modeling and Management for Natural Disasters: Innovation by the Numbers Presentation
Thanks for sharing, Pegah. I watched the video. Some takeaways follow, that might be good prompts for discussion.
1. The description of Risk Science left something to be desired. As practitioners, do you have your 20 second answer?
2. Risk can and will deteriorate over time. We often assume stationary, but need to take a lifecycle view.
3. Complex systems may be non-intuitive and proper insight can only come from mathematical modeling. Does anyone have a good example to share?
4. The ship allision with the Francis Scott Key bridge was ascribed as a black event. Is this really the case, since the data shows this to happen globally about every 18 months. https://apnews.com/article/bridge-collapses-barges-list-1f2d6261d523ddc625aaaf3b32c626bc
5. The use of incentives was mentioned as a mitigation. Does anyone have a good example?
6. The issue of risk normalization was raised in terms of human behavior misconceiving lack of an event as evidence that the event is very rare. I think the tragedy that took place in the Texas Hill Country this past July is a prime example of risk normalization. What other prominent examples do you have and why?
7. The over reliance on modeling and importance of common sense thinking was cited. The counter argument was also made that some complexity is beyond common sense. This is a great tension to explore more and keep at the forefront of our discussions.
------------------------------
Mitch Winkler P.E.(inactive), M.ASCE
Houston, TX
Original Message:
Sent: 10-15-2025 12:43 PM
From: Pegah Farshadmanesh
Subject: Risk Modeling and Management for Natural Disasters: Innovation by the Numbers Presentation
This presentation was interesting: Risk Modeling and Management for Natural Disasters: Innovation by the Numbers. The recording can be found here: https://www.nationalacademies.org/event/45862_10-2025_risk-modeling-and-management-for-natural-disasters-innovation-by-the-numbers
------------------------------
Pegah Farshadmanesh Ph.D., M.ASCE
Memphis TN
------------------------------