Abstract¶
OACT is an open-source, public-data decision-support prototype designed to help stakeholders detect, investigate, and quantify infrastructure disruption risk during extreme weather events (e.g., flooding and atmospheric river scenarios). This talk presents a reproducible “Right to Replicate” architecture: public data ingestion, auditable transformations, transparent scoring, and an evidence-first workflow designed for responsible reuse by universities and public-sector partners.
We will walk through the end-to-end pipeline (signals → features → risk scoring → recommended actions) and highlight design choices that improve reproducibility and governance, such as dataset lineage, assumptions logging, and decision traceability. By focusing on “auditability by default,” OACT addresses the critical gap between opaque proprietary risk models and the need for transparent, accessible tools in public-sector resilience planning. Attendees will leave with a concrete pattern for building analytics prototypes that bridge research, open infrastructure, and real-world decision needs.
Yidan (Lena) Hu | Independent Researcher¶
Yidan (Lena) Hu is the Lead Data Scientist for the Open Analytics Control Tower (OACT), where she leads the design and implementation of the system’s risk modeling framework, focusing on translating compound environmental signals into predictive, decision-ready indicators of infrastructure disruption. In her professional career, Lena is a senior-level data scientist in a top-tier tech company, specializing in large-scale analytics and machine learning systems within the media and entertainment industry. Her current independent research establishes a reproducible, auditable methodology for operational risk intelligence, directly supporting public-sector resilience efforts and aligning with federal and state priorities for critical infrastructure protection. Lena holds a Master’s degree in Business Analytics from Fordham University.
Laisi (Maggie) Ma | Independent Researcher¶
Laisi (Maggie) Ma is a tech leader in a top-tier tech company, dedicated to cloud technology, AI adoption on cloud, and open source contribution. She has contributed to the open source project AWS JS SDK V3.
Hao He | Independent Researcher¶
Hao He is the Data Engineering Lead for the Open Analytics Control Tower (OACT), responsible for designing and building the multi-source ETL pipeline that transforms raw government signals into county-level risk intelligence. Drawing on experience in large-scale data engineering, finance analytics, and operational excellence at a top-tier technology company, he specializes in integrating heterogeneous public data streams including USGS hydrology, ERA5 atmospheric reanalysis, and NOAA forecasts into reproducible, auditable analytical workflows. His independent research on OACT focuses on making open environmental data operationally useful for supply-chain resilience, aligning with federal and state preparedness goals under S.257. Hao holds an M.S. in Business Analytics from the University of Texas at Austin and a B.S. in Economics and Mathematics from the University of Michigan, Ann Arbor.
Yuan-Jiun (David) Sung | Independent Researcher (Principal Investigator, remote) | OACT Project Hub¶
Yuan-Jiun (David) Sung is the Principal Investigator and Lead Architect for the Open Analytics Control Tower (OACT). A senior data professional with over a decade of experience architecting large-scale intelligence systems for top-tier technology and financial institutions, he specializes in translating complex environmental signals into actionable, auditable risk decisions. His current independent research aligns with federal and state resilience initiatives, providing open, reproducible infrastructure to protect critical supply chains and “Missing Middle” operators from climate-driven failures. He holds Master’s degrees in Software Management and Electrical and Computer Engineering from Carnegie Mellon University and Purdue University.