v6.4

What's Changed

• Improved facility locations
  • OSM tag filtering has been improved to reduce number of wrongly assigned facility types
• Update work location assignment within Berlin
  • Work locations now use weighted sampling during location choice
  • The weight is based on an attraction factor and zone specific probabilities determined from survey data
• Updated GTFS schedule
  • The schedule is now based on the 2024-11-19
  • Note that there are major changes how stops and links between them are created
    • Stops having the same gtfs parent_id and route types are merged together, allowing agents to find better PT connections
    • The PT network is created with loop links (on each PT stop) instead of duplicating stops, which also improves connections
    • The Berlin Ringbahn is manually adjusted so that each train drives multiple loops the whole day
• New income calculation
  • In previous versions income was used directly as household income from the survey data
  • Now, the income is calculated as personal equivalent income, which is the household income divided by equivalent household size
    • See https://en.wikipedia.org/wiki/Equivalisation
  • Corresponding attributes have been added to the population file
• Bike mode updated and recalibrated
  • Bike is now routed on the network, which results in more realistic travel distances
  • Bikes are not simulated on the network, and no link events generated yet. This will likely be added in future versions.
  • The road network includes bike infrastructure and corresponding attributes
  • The bike infrastructure is not fully complete yet, and has to be carefully evaluated first for bike centric studies
  • The avg. bike speed has been set to match SrV2018 survey data (~10.3 km/h)
• New dashboards
  • PT Transit viewer
  • Emissions
  • Noise

Full Changelog: v6.3...v6.4

v6.3

What's Changed

• Include additional trip analysis and updated dashboard
  • Mode share is now analyzed by age, income, employment, economic_status
• Updated population, which now include reference modes for certain persons
• The reference modes represent modes that a person has actually used in reality
• Allows to evaluate the quality of the model in terms of mode choice
• ReplanningAnnealer module is now activated by default

Full Changelog: v6.2...v6.3

v6.2

What's Changed

• Updated network and gtfs schedule
  • Network is now based on late 2022 osm data
  • PT schedule is based in mid 2023, earlier data of this feed was not available
  • The network conversion now uses a SUMO converter, a microscopic representation of the network will be available as well
  • Free speed of the network has been calibrated as described in the paper "Road network free flow speed estimation using microscopic simulation and point-to-point travel times"
• Walking speed increased to match reference data
• Recalibration of mode constants (ASCs)
• Updated commercial traffic, especially the trip durations should be more realistic
• Added a run class for DRT scenarios
• Please note that the 6.3 version is released nearly at the same time
  • Because of that 6.2 was not calibrated to the same accuracy
  • Rather use the newest version, which also includes all features of 6.2

Full Changelog: v6.1.1...v6.2

v6.1.1

What's Changed

• Fix ASCs in tbe 6.1 config
  • The calibrated ASCs have not been correctly copied to the 6.1 config, this is now fixed.
• All input files remain the same
  • The existing output remains the same as well, because this run was using the correct ASC values

• drt for berlin v6.x by @tschlenther in #123
• update pt-extensions to latest commit (b39f03f2f0 from 2023-12-07) by @tschlenther in #124

Full Changelog: v6.1...v6.1.1

v6.1

What's Changed

• Trips from survey data are filtered slightly different

  • Previously, trips that did not have the flag E_WEG_GUELTIG, have been ignored.
  • However, this flag is not considered in SrV reports, or the Berlin website that reports the modal share.
  • To be more consistent with these reports, these trips are now considered as well.
  • This leads to slightly different mode shares compared to the previous version.

• Commercial traffic was updated to include recent changes

  • Temporal distribution was updated according to survey data
  • Separated car and freight/truck mode

Full Changelog: v6.0...v6.1

OpenBerlinScenario v6.0

Release for the OpenBerlinScenario.

• Compared to the pre-release, costs config and calibration parameter have been updated

v6.0-parallel-qsim-benchmark

This release contains the scenario which was run to compare to the Paralle Qsim Rust implementation.

This release is not meant to be used by anyone else. Use the regular v6.0 release instead.

OpenBerlinScenario v6.0-pre

This is a pre-release that can be used to test the functionality of the scenario and try new studies / policy cases. This is not the final calibrated model, some parameters are going to change.

Mosaik-2-01

This release captures the run class used for the simulation experiments conducted within the mosaik-2 project, so that this snapshot can be referenced with a doi

Janek's release

I need to run the scenario once again, but the dependencies underneath have changed. Again, creating a tag, in the hope of a passing ci build.