What the discipline has produced.
Twenty-four years of right-fidelity modeling across energy, environment, and public health — the same discipline now aimed at neurodegeneration.
← The MissionBy Michael Key · ORCID
Who built this, and what has the work produced.
Background
Twenty-four years at the intersection of math, simulation, and high-stakes decisions — from MITRE’s intelligence analysis center to major aerospace contractors to running an independent practice. The credentials and the arc that leads to this mission.
Read the background →The method on other problems
Thirteen studies on real, high-stakes problems — energy grids, groundwater, public health, a longevity cohort, higher education — most showing that right-fidelity modeling with honest uncertainty holds up, and on the highest-stakes ones, changes the decision. The evidence that the discipline now aimed at Alzheimer’s and ALS carries weight.
Browse studies ↓The method, proven on real problems.
None of these thirteen studies is about neurodegeneration — and that’s the point. Each is evidence that the discipline — calibrated, honestly-bounded modeling — holds up on real, high-stakes problems, and often changes the decision. The studies closest to the mission come first; the rest is the same method, applied across other domains.
Human-cohort & health studies
The method on real human-health data — the nearest neighbors to the Alzheimer’s and ALS work. New studies aimed closer to cognition and aging will join this group as they’re finished.
What predicts how long we live?
64,000 real people across the NHANES and HRS cohorts, 13 investigations. ML does better than domain knowledge on some questions; domain knowledge does better on others. The right method depends on the question — the same judgment the neurodegeneration work turns on.
Longevity study →California freight & ozone
57 chained investigations on a 700K-person cohort. Key finding: cutting NOx raises ozone in 99.95% of scenarios — complicating the common assumption that cutting NOx always lowers ozone.
RFAQ study →The method on other domains — eleven studies · energy, water, environment, emergency response, education
Can the grid absorb the data-center surge?
Can PJM’s grid absorb 40 GW of new data center load? 200-draw Monte Carlo, 4 chained questions. Validated against FERC/NERC filings.
PJM study →Military base groundwater
Screening model says pump-and-treat is cheaper. Full physics analysis reverses the decision. Physics equations and Monte Carlo, not neural networks.
PFAS study →Evacuation & insurance
Four real wildfires, NASA FIRMS satellite data, USGS elevation, NIFC perimeters. Right model shifts evacuation triggers 1–7 hours earlier across 16 communities.
Wildfire study →Can the river keep its promises?
Legal over-allocation vs. physical supply. 12 investigations. Global sensitivity shows the delivery obligation drives just 7.3% of breach risk — inflow drives 92.9%.
Colorado River study →Offshore wind & right whales
Pile-driving noise propagation vs. NARW critical habitat. 8 investigations, 3 sites, real SSP profiles. Coordinated scheduling cuts total whale takes 3.3× and creates 9 quiet hours/day — at zero added cost.
Ocean acoustics →Texas grid resilience
Five questions about the grid that failed in Winter Storm Uri. Physics-based dispatch, investment tradeoffs, location sensitivity.
Texas energy →Where should a data center go?
Where should a large data center go when transmission is the binding constraint? Spatial optimization over grid physics and siting cost.
DC Siting →Where do the charging gaps fall?
Where do the coverage gaps fall, and how big should each station be? Network planning under grid constraints, with battery-storage tradeoffs.
EV Charging →How much fidelity does the decision need?
3 fires, 4 fidelity levels, 200 Monte Carlo draws. Four viability gates, 2 interactive tools. How fidelity affects the insurance decision.
Wildfire insurance →College closure risk
4,000+ institutions modeled with real College Scorecard and IPEDS data. Six investigations, three model tiers, three interactive tools. ML vs. rules-based risk scoring on a question where the real outcome is known.
College closure →Is your degree worth it?
65,935 college programs analyzed with real IRS earnings data. 34% have negative lifetime ROI. Data analysis, not modeling — but the same question-first discipline applied to a hard, consequential question.
College ROI →