This document tracks future enhancements and features that are out of scope for the current MVP but planned for later.
- Goal: Use live data and historical patterns to validate transfers more intelligently than a fixed buffer.
- Implementation:
- Platform Awareness: Use
TimetableServiceto get real-time platform info. - Dynamic Buffer:
- Same Platform: 2 mins
- Same Station: 7 mins
- Cross-City: 10 mins
- Delay Propagation: If incoming train is delayed, check if the connection is still reachable.
- Platform Awareness: Use
- Goal: Use the downloaded Parquet files (
datachaos/deutsche-bahn-data/) to predict delays for future dates. - Implementation:
- Ingestion Script:
server/scripts/ingest_delays.pyto populatedelay_patternstable. - SimulationService: Query this table to provide "Probable Delay" for future trips.
- Ingestion Script:
- Goal: Replace the static GraphService with a time-dependent routing algorithm.
- Implementation:
- CSA (Connection Scan Algorithm): Implement CSA on the raw GTFS
stop_timesdata. - Raptor: Alternative algorithm for multi-criteria optimization (e.g., "Fastest" vs "Fewest Transfers").
- CSA (Connection Scan Algorithm): Implement CSA on the raw GTFS
- Dockerization: Containerize the application for easier deployment.
- Caching Strategy: Move from JSON/GML file cache to Redis or similar for the Graph.