Operating large journey planners and supplying data in the internet is done in processes. The work flow can essentially be divided into
- Create and collect data
- Check data
- Convert data
- Distribute data
Journey planners that serve areas with millions of commuters need to be able to process large amounts of data. With the emergence of real-time systems, transport data need to be updated on a daily basis. The involved processes need to run quickly and often. In the journey planning areas of the English Traveline Regions Greater London, South East, East Anglia, East Midlands, West Midlands and South West there are more than 30 million people. The public transport there includes approx. 20,000 routes. The system requires new data every day.
- Link South East
Currently, the largest system in German-speaking regions, operated by the NVBW (Nahverkehrsgesellschaft Baden Württemberg), includes data from the Federal German States of Bavaria, Baden Württemberg, Rhineland-Palatinate, North Rhine-Westphalia, Bremen, Lower Saxony, Saxony, Mecklenburg-Western Pomerania, as well data in Austria, Switzerland and the border region of French Alsace. The entire region has more than 70 million people and the various public transports systems have more than 63,000 routes.
The quality of the planned journeys can no longer be comprehensively tested due to the volume of data, but the process itself must include quality controls. High-quality data are a prerequisite for premium journey planning. We understand quality to mean
- Referencial integrity (serviced stops must exist)
- Plausibility (vehicles cannot exceed certain speeds)
- Integrity (all public transport services must be displayed)
- Up-to-dateness (services need to be duly added to the system at the right time)
In addition to timetable data a journey planner also includes geographical data. The System must be able to:
- Import geographical data
- Reference timetable data
- Supply the journey planner with geographical data to enable routing and mapping
- Calculate map tiles
The final requirement for a door-to-door journey planning system is address data. This, along with stops, localities and POIs must be made available to provide origin and destination identification.
The data management system has the following tasks
- Enable and support a fully-automatic process to supply the journey planner
- Execute constant quality controls in all processes and documentation of the quality.
DIVA data are processed in subnets that are integrated into a journey planner. In the NVBW example above, more than one hundred subnets need to be processed. The following image shows procedures for 3 subnets.
Parallel work is essential for fast processing.
The sytem constantly checks the data and reports the results to the task manager and notifies suppliers.
Logs and a bar graph display the progress of automized processes. The following bar graph has increments in hours to the left and minutes to the rights.