Skip to content

Release Notes

This is a summary of the key changes in subsequent patches for version 23 of the Aimsun Next environment.

23.0.1

Improvements

Path assignment plan

  • For a dynamic DUE simulation loading as input a path assignment plan with a single, multiple or mix of either static or DUE path assignment items.
  • If the simulation period of the dynamic simulation (either SRC or DUE) loading the DUE path assignment plan and the simulation period of the DUE path assignment file are not consistent in terms of start time and duration, then the simulation will be cancelled.
  • The path assignment plan can be loaded as input in a DUE simulation even when route choice intervals between the experiment of the input DUE path assignment item(s) and the dynamic experiment don't match. In that case, the percentages of the OD paths assigned in the dynamic experiment use proportion of the intervals of the DUE path assignment. For example, if the path assignment file generated in a DUE with a demand from 08:00 to 09:00 with 10 minutes route choice interval (in total 6 intervals) and the dynamic experiment that loads this DUE path assignment plan starts at 08:00 and finishes at 09:00 with 15 minutes interval (in total 4 intervals), then the first interval (08:00-08:15) will use 66.6% of the first interval in the path assignment (08:00-08:10) and 33.3% of the second interval in the path assignment (08:10-08:20).
  • In the path assignment result are now displayed the correct used paths per interval with the correct percentages. If the user wants to use a one shot simulation to reproduce the result of a DUE simulation, it is now necessary to configure a dynamic route choice model because otherwise it merges all paths to a unique interval (before it could configure the experiment with stochastic route model "Fixed Using Travel Times Calculated under Free-Flow Conditions").
  • The warm-up period is not considered in the path assignment plan items.
  • When the a path assignment item generated for a period (i.e. 08:00-09:00) is reduced manually inside the path assignment plan (i.e. 08:30-09:00), there is an option to select which interval (i.e. 15min) of the DUE path assignment item should be picked by the dynamic simulation (i.e. 08:30-09:00). In this example, the user chooses to start from 08:00.

Statistics

  • Mean Queue in section output have changed from an average of vehicles in state of queue by lanes to an average of vehicles in state of queue weighted by lanes length.
  • A new Pedestrian Crossings output has been added: Pedestrian Maximum Waiting Time: Maximum calculated time pedestrians are waiting at the Pedestrian Crossing (in seconds) for the pedestrian traffic light to become green.
  • A new Transit Stop output for pedestrians has been added: Pedestrian Maximum Waiting Count: Maximum calculated time pedestrians are waiting at the Pedestrian Crossing (in seconds) for the pedestrian traffic light to become green.
  • Statistics for Groupings and Grouping Categories can now be refreshed for more that one originator at once.

Control plans

  • A Node filter has been added to the Master Control Plan editor, so that it is easier to identify which control plans apply to a specific node.

Static Assignment

  • Transit Volume is now accessible in all Cost Function evaluations through the GKFunctionVolume methods getVolume(...) and getVehicleVolume(...). The GKUserClass associated to GKVehicleType (f.i. Bus) in a Transit is the one with that vehicle type (Bus) and a null purpose.
  • Convergence Output tab does not longer appear for Stochastic Assignments as the concept of Convergence does not apply for this type of assignment

Mesoscopic Network Checker

  • As a Fast Track Feature for version 23.0.1, there is a new option for Queue Jump Gridlock Prevention in the Mesoscopic Network Checker, that will look for vehicles stucked in a queue waiting for a non-congested turn behind vehicles that wait for a congested turn and let them, after some time, jump the queue.

Microscopic API

  • Function AKIODDemandGetDemandODPair now returns the real simulated demand and the return value is an integer instead of a double.

Surrogate Safety Assessment Model

  • To improve the precision in the trj files, all the coordinates will now be translated closer to the origin of coordinates. The translated distance will be stored in a json file with the same name.

Bug Fixes

DUE simulations

  • Path Update After Virtual Queue is no longer supported in DUE simulations. The usage of these option introduced inconsistencies in the calculation of link costs and rgap.

Hybrid Macro-Meso

  • Close turn actions with specific lanes were closing the turn. These turn close actions with specific lanes are not considered now in the macro area.
  • Crash when the close turn action was referring to a non-existent turn has been solved.

Four-step Model Experiment

  • Now it is possible to connect from a script box back to the loop box, provided the slots match.