JOR 3.1: Python/PyMC dashboard for UAP collision risk index
If you're interested, I built a Python module for the JOR V3.1 framework that uses a vectorized PyMC sampling pipeline to calculate a Collision Risk Index (CRI). It basically treats physical presence certainty as a baseline and uses kinetic flight anomalies (like instantaneous acceleration) as a multiplier:
CRI = SOP_Mean * (1 + Flight_Mod)
The outputs map into standard hazard tiers (Low, Elevated, Critical) and feed into a Streamlit dashboard I set up, which plots the track relative to safety limits and MCMC uncertainty ranges.
The main goal is just to give aviation stakeholders a standardized metric for threat mitigation.
Full methodology and report are archived here if you want to dive into the math (the open-source repository code link is included in the Zenodo description):