Tidy Transit: Real Life Data Modeling for Public Transportation
summary
An open source data warehouse ingests schedules nightly and vehicle positions every ~20 seconds for ~800 agencies (200 fixed‑route) so statewide analyses can identify a Route 44 that averages 5.5 mph in afternoon peak, and produce per‑agency GTFS quality reports. Payments arrive ~2 days via EMV payments and feed merchant dashboards and fare‑capping stats. The stack is Airflow, Kubernetes, BigQuery with JupyterHub + siuba notebooks; a mobility marketplace and pre‑negotiated procurements enable contactless tap pay that already works in Sacramento.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 3:20Cal-ITP mission
- 5:08Trip planning with real-time data
- 7:59Standardizing EMV payments
- 10:48Equity-focused fare discounts
- 11:33Data standards for transit analytics
- 14:02Speed analysis for transit investment
- 15:30Q&A
related talks
16mTidy Transit: Real Life Data Modeling for Public Transportation
10mSmart Commuting
43mHacking your local bus system for the greater good
48mSwiss Transport in Real Time: Tribulations in the Big Data Stack
30mCreate Detailed Traffic Models and Simulations
36mNext generation plug & play IT-systems for public transport
46mSmart Transit: Real-Time Transit Information with FLaNK
31mServerless datastream processing