Bill, thank you for your thoughtful guidance. Since this discussion pushes beyond conventional boundaries, I turned to the ASCE Continuing Education catalog - specifically the course "Innovations in Port Engineering: AI, Digital Twins, and Structural Health Monitoring (OTIPC2520)", which I completed recently.
This course provided a consolidated technical foundation that ties together AI‑enabled inspection, digital twin methodologies, and structural health monitoring for waterfront and maritime infrastructure. It also pointed me toward the deeper technical material presented at Ports® '25, where Jacobs, SGH, VCS Engineering, and HydroMapper demonstrated the most advanced inspection technologies currently deployed in the field.
Building on that foundation, one of the strongest industry benchmarks we currently have is the Jacobs work presented at Ports® '25, where LiDAR, multibeam sonar, photogrammetry, and UAV/ROV platforms were combined to produce high‑density datasets for waterfront risk evaluation. These are indeed successful applications - but they also reveal a deeper structural gap that aligns with the concern you raised earlier.
Even in these high‑tech deployments, digital twins remain state‑based representations, not continuous risk engines. The Ports® '25 handouts repeatedly highlight this limitation:
– LIDAR and sonar generate massive point clouds, but the data is episodic.
– GPR, impact‑echo, and corrosion potential provide critical deterioration signals, but they are not streamed.
– IoT devices can monitor usage and incidents, yet their integration into unified risk models is still fragmented.
This is the same bottleneck I described in my earlier post on dam safety - the absence of a continuous structural pulse capable of merging heterogeneous data streams into a single operational ecosystem. My recent work with autonomous workflows attempts to bridge this gap by orchestrating live sensor feeds, NDT outputs, and remote‑sensing datasets into a unified RAG‑driven environment that maintains both engineering fidelity and operational scalability.
If you feel this convergence could benefit our ongoing discussions within the CoP, I would be glad to prepare a focused technical note or a short case‑based summary that aligns with the needs of our community. Your guidance on the most useful angle or depth would be highly appreciated.
Best regards,
------------------------------
Abubakr Gameil, R. ENG, M. ASCE®️,
MSc-Holder, [ SEI, EWRI, CI, ISSMGE ]Mermber
Past / Chairman & Director General
Almanassa Engineering International Co. Ltd,
Khartoum, Sudan
Currently / UAE- Humanitariam Residency
NXN- Central branch -Al Fujairah,
PO.Box : 1142 (Fujairah)
------------------------------
Original Message:
Sent: 06-25-2026 11:09 AM
From: William McAnally
Subject: Application of AI for Risk Assessments
Thanks, Jameel. That was mostly beyond my ability to understand, but I'm sure many in our community will find it useful.
Links to articles and reports, particularly case studies, will be helpful to my understanding.
Bill
------------------------------
William McAnally Ph.D., P.E., BC.CE, BC.NE, F.ASCE
ENGINEER
Columbus MS
------------------------------
Original Message:
Sent: 06-23-2026 12:42 PM
From: Abubakr Gameil
Subject: Application of AI for Risk Assessments
Thank you Bill, I promise I will. Actually there are three Initiatives ( Refugees replacement-2023, Feeding Aid - (2024), and statistic programe to support people after crises - (2025) respectfully, All of these had been an AI-based with deferent types of machine learning.
The first one based on Automation with analytic AI- Jason code ,using (Microsoft Form, Excel, Power Automate, powerBI) tools,
The data sets were, Qualitative, Quantitative and spatial.
The second one, was based on Analytic -AI mainly using synchronized spread sheet and PowerBI, the data were also came from deferent nature (qualitative, Quantitative and spatial )
The last one was more complexity , It was built on LLM/ RAG machine learning, Low code AI ,Automation, using microsoft form, spread sheet,power Automate, AI-Builder, PowerApp ( to build smart UI) and PowerBI for analysis.
Bill, really I do not know know what is the right context should I elaborate each one, If you have any suggestion I would be happy to see.
Thank you very much and welcome
------------------------------
Abubakr Gameil, R. ENG, M. ASCE®️,
MSc-Holder, [ SEI, EWRI, CI, ISSMGE ]Mermber
Past / Chairman & Director General
Almanassa Engineering International Co. Ltd,
Khartoum, Sudan
Currently / UAE- Humanitariam Residency
NXN- Central branch -Al Fujairah,
PO.Box : 1142 (Fujairah)
------------------------------
Original Message:
Sent: 06-23-2026 11:35 AM
From: William McAnally
Subject: Application of AI for Risk Assessments
Wow! That sounds impressive, Jameel. The LinkedIn post is intriguing. I hope we can hear more about your team's work here in the CoP.
Bill Mc
------------------------------
William McAnally Ph.D., P.E., BC.CE, BC.NE, F.ASCE
ENGINEER
Columbus MS
------------------------------
Original Message:
Sent: 06-23-2026 04:31 AM
From: Abubakr Gameil
Subject: Application of AI for Risk Assessments
Hi Mitch,
Thanks for throwing these great questions into the mix. It's the perfect time to look at what's actually happening behind the "black box" of generative AI. LLMs don't understand engineering intent; they simply predict the most statistically probable next token. This explains the exact frustration Bill pointed out-why standard AI often just echoes back what you want to hear. The output is entirely trapped by the immediate context it's fed.
To solve this, I use a Retrieval-Augmented Generation (RAG) architecture within a closed environment, giving the AI access only to a verified reference library. I deploy this natively using Power Apps and Power Automate (AI Builder).
I recently stress-tested this exact low-code framework under extreme constraints in a high-stakes crisis environment. Operating within a strict 72-hour execution window "For more information; kindly see my LinkedIn post , we used it to handle complex, multi-type data influxes, real-time mapping, and automated sorting for thousands of data points. The system held because we tightly controlled the hyperparameters to maintain absolute data integrity:
Temperature (Creativity): Dropped to the lowest quadrant. This forces the AI to be strict and deterministic, sticking 100% to verified reference files instead of being "creative."
Precision (Top_p): Capped below average to restrict wild logical leaps and completely prevent "hallucinations."
Whether you are managing emergency logistics or structural risk assessments, the professional workflow remains identical:
Isolate the Repository: Lock down your clean codes, specific contracts, or standard procedures (your RAG library).
Target the Prompts: Instruct the AI to map and extract risk types, levels, and consequences only from those source files.
Calibrate Thresholds: Lower both temperature and precision settings.
Validate: Run rigorous testing loops inside that closed library before relying on the output.
Recommendation:
Use this setup strictly as a first pass during the initial risk identification phase to stretch your thinking and catch blind spots. It is a powerful assistant for handling massive data, but it is not an autopilot. You absolutely still need a highly experienced team of risk professionals at the wheel to audit the system and make the final, critical action.
Best regards,
------------------------------
Abubakr Gameil, R. ENG, M. ASCE®️,
MSc-Holder, [ SEI, EWRI, CI, ISSMGE ]Mermber
Past / Chairman & Director General
Almanassa Engineering International Co. Ltd,
Khartoum, Sudan
Currently / UAE- Humanitariam Residency
NXN- Central branch -Al Fujairah,
PO.Box : 1142 (Fujairah)
------------------------------