Parameters of the simulation-based dynamic assignment
Simulation-based dynamic assignment: Basis tab
Element |
Description |
Use current assignment result as initial solution |
If this option is checked, the assignment will use existing paths and volumes of the previous assignment as an initial solution. Note This option is recommended in the case of modified demand (e.g. due to 'Read matrix additionally' or matrix editing in the matrix editor). First, the former assignment result is scaled according to the new demand and then the assignment continues to calculate the new result. Usually, this requires fewer iterations than a complete recalculation. In this way, a lot of computation time can be saved in models with feedback from the assignment calculation to the demand calculation. Iteration offset for volume balancing If an existing assignment is used as initial solution, you can enter an iteration offset to redistribute less volume. The iteration offset o ensures that volume is redistributed in such a way as though o iterations were calculated already. In Iteration n, 1/(n+o-1) of the demand will then be redistributed. |
Termination condition |
The procedure is finished when either the maximum number of iterations has been reached or the gap value goes below the defined threshold. Maximum number of iterations Here, specify the maximum number of iterations after which the procedure ends if no convergence occurs. Maximum gap Measure of the deviation in the iteration Maximum number of additional iterations Enter the maximum number of iterations that shall be calculated additionally. Note Convergence is only considered achieved in SBA if all vehicles leave the network in the last iteration. If the maximum number of iterations is calculated, but there are still vehicles in the network in the last iteration, further iterations can optionally be calculated until all vehicles exit. As soon as all vehicles in an iteration have exited or as soon as the maximum number of additional iterations is reached, the assignment is ended. The gap is not considered as a convergence criterion in the additional iterations. |
Simulation |
Random seed In the case of otherwise unchanged settings, you can determine a different result with a new random seed. The random seed applies to random generators in the simulation and to the departure time choice. Note In the directory C:\Users\Public\Documents\PTV Vision\PTV Visum 2024/Examples, you can find an example of use on this topic. The SBA MultiRun example describes an application in scenario management for multiple executions of SBA with different random numbers. Use gridlock avoidance If gridlock avoidance is used, cycles are identified during the simulation that are about to congest completely. At the nodes of an identified cycle, the priority rule is temporarily changed so that vehicles in the cycle have priority before vehicles outside the cycle. In many cases, this prevents the cycle from being entirely congested. If, in this situation, the foremost vehicle on each link of the cycle wants to move to the next link of the cycle, then no vehicle can continue because all links are full. The result is a gridlock that can no longer be resolved in the current simulation. This happens, for example, at incorrectly modeled roundabouts. These are roundabouts that are not modeled as nodes with control type 'roundabout', but consist of several nodes with short links in between, and where the vehicles have no right of way in the roundabout. This is the case, for example, if roundabout nodes have the control type 'unknown'. If enough vehicles have left an identified cycle, the original priority rule is restored. Information on the use of gridlock avoidance can be found in the message window after the assignment. |
Paths |
Limit number of path searches If this option has been selected, you can set a last iteration for the path search. Last iteration for path search Enter the last iteration, in which new shortest paths shall be sought After this, only volumes between paths of the same OD pair will be balanced in the iterations. No more new shortest paths will be sought. Minimum path volume Here you can define a lower limit for the volume of a path. Notes During volume balancing, volumes for the path are set to 0 if they fall below the specified value. This can considerably reduce the number of paths found. Values that are too high can impair the convergence of the assignment. Time interval duration for averaged travel times During the simulation, mean travel times will be measured on time intervals of this length. Travel times in these time intervals will be used for the shortest path search and for the determination of path costs (experienced travel times). |
Simulation-based dynamic assignment: Assignment time period tab
On this tab, define the assignment time period and sub-divide it into smaller time intervals for calculation. The time intervals can be of different sizes, for example, it is possible to use longer time intervals during nighttime and shorter ones during peak hours. For that purpose, sub-divide the assignment time period into several larger time periods and then define the length of the time intervals by period.
Element |
Description |
Assignment time period |
Enter the desired assignment time period. from Start of the assignment time period Note The assignment time period can be longer than 24 hours so that an appropriate post-assignment period can also be defined for a 24h assignment. to End of the assignment time period Post-assignment time period Enter the time period in which vehicles still pass through the network but no new demand is loaded anymore. Note The post-assignment time period should be large enough to allow all vehicles to exit the network by the end of the simulation. |
Time intervals for balancing |
Use the buttons below the table to add or delete entries. Within the time intervals, routes are sought in the assignment and volumes are balanced. Notes Only within the balancing time intervals, demand is loaded into the network. Paths have a demand per balancing time interval during the assignment and these volumes between all paths of an OD pair are balanced separately within each balancing time interval. |
Use this button to create analysis time intervals for the specified time intervals. |
Simulation-based dynamic assignment: Volume balancing tab
Element |
Description |
Method for volume balancing |
Select the method for the distribution of the demand to the paths (Fundamentals: Distribution of demand across paths (volume balancing)). MSA Averaging according to Method of Successive Average (MSA) Cost-proportional Cost-proportional volume balancing |
Step size control |
Here you can set the method for the step size control (Fundamentals: Distribution of demand across paths (volume balancing)) MSA If the option has been selected, 1/n of the volume will be redistributed in iteration n. Note This option is suitable for short assignment intervals. Time interval-dependent If the option has been selected, the volume is redistributed up to the iteration specified under Start with time interval-dependent step size after iteration, just as with option MSA. After that, the step size is determined depending on the balancing time interval (Fundamentals: Distribution of demand across paths (volume balancing)). Note This option is recommended in models with long assignment time intervals, where several gridlocks might occur distributed over the assignment time period. |
Simulation-based dynamic assignment: Base volume tab
Optionally, a road-related base volume can be taken into account in the procedure, for example, to include the volume of public transport.
Element |
Description |
Use base volume on links |
If this option is selected, base volume from a link attribute is taken into account in the assignment. Select the required link attribute and enter a factor for the base volume. The base volume can vary over time in the assignment period and requires the allocation of a transport system. Notes If you select an attribute with subattribute <time intervals>, the base volume is time interval-dependent. For all other attributes, the number of vehicles per hour applies. Base volume vehicles are considered in all simulation-based evaluations (simulation number of vehicles or simulation vehicle hours traveled t0 in the assignment quality list, for example, and SBA density or SBA utilization at network objects). |
Transport system |
In the case of a base volume, additional vehicles of the specified transport system enter the simulation in the demand period. Select a transport system for the link base volume. |
Simulation-based dynamic assignment: Macro-meso simulation tab
You can specify areas in which mesoscopic simulation is carried out in simulation-based assignment(Fundamentals: Macro-meso hybrid simulation).
Element |
Description |
Use macro-meso hybrid simulation |
If the option is selected, vehicles are simulated only on selected links in the meso area. You define this area via the attribute selected in the Simulate link mesoscopically section. If the value of the attribute selected here for a link is not 0, the link is part of the meso area. On all other links (macro area), vehicles travel at the static speed resulting from the VD functions. |
Simulation-based dynamic assignment: Visualization tab
During the simulation-based dynamic assignment, vehicle trajectories can be exported for visualization purposes (Visualizing vehicles of the SBA assignment).
Element |
Description |
Record vehicle trajectories for visualization |
If the option has been selected, vehicle trajectories are recorded in the last simulation. Restrict trajectories to active links Select this option to record vehicle trajectories for active links only. Restrict trajectories to time period Select this option to record vehicle trajectories for the specified time window only. Specify the desired time period. |
Temporal resolution |
Input of the temporal resolution for the vehicle trajectories |