Python flying at 40,000 feet
summary
Shows a working pipeline that ingests black‑box HDF data, segments flights and detects the end‑of‑takeoff by modelling fuel‑flow and EGT around a 20‑second window tied to V‑NAV. Uses Python (numpy scipy pandas pymc), Jupyter and R for exploration, stores results in FlightDataAnalyzer and relies on 4,000,000 flights for priors. Implements a Bayesian posterior over end‑of‑take‑off (assumes parameter independence), runs pymc MCMC with 10,000 samples discarding an 8,000 burn‑in, and omits failed sensors rather than failing the flight.
AI-written summary · about
chapters (7)
- 0:00Flight data monitoring intro
- 3:18Non-punitive monitoring necessity
- 7:23How to become a flight data analyst in one minute
- 15:09Detecting takeoff end with Python
- 23:43Building model with PyMC
- 33:50Predicting takeoff end
- 37:43Data sharing challenges
related talks
42mMaking Sense Out of Flight Test Data with Python
18mSit back, relax, and enjoy the talk: a journey through FlightRadar24 with Python
43mRemove before flight: Analysing flight safety data.
35mTouchdown Localisation with aircraft flight data
45mInteractive Analysis of (Large) Financial Data Sets
30mOn the Fly and On the Flight: Scientific Data Analysis Beyond the Beamline
22mTaking Flight with AI and ML: Revolutionizing Airline Ops
43mPython in the Sky: In-Flight Entertainment with Python