Data flow chart
Fig.1: Flow chart of DLSim-MRM
DLSim-MRM consists of 5 main modules as follows:
DLSim-MRM uses different types of input data depends on desired simulation type which include static or static+dynamic. If users want to use only static mode which include user equilibrium assignment mode, they should present the files in the blocks a and b.
Network data block has 2 main input files as node.csv, link.csv.
Fig.2: node.csv
Fig.3: link.csv
Demand and setting data block has 2 main input files as demand.csv, settings.csv.
Fig.4: demand.csv
Fig.5: settings.csv
Scenario data block has 3 main parameters as work zone, incident, information in scenario.csv.
Fig.6: scenario.csv
link.csv Signal timing data block has 4 main parameters as cycle time, start and end of green time for each link with movement text id (mvmt_txt_id) n link.csv. ** add the signal file picture.
Fig.7: link.csv
(macroscopic) routing module uses the queue-based VDF to compute link volume, travel time and outputs the static MOE in file static_link_performance.csv.
Based on the first-stage routing that uses static VDF based estimates as initial values for travel time, DLSim can use Lagrangian relaxation-oriented price to modify the generalized cost function to reflect the real time information changes.
**Key point: Understand the connection and recognize potential inconsistency between different resolutions **
Needs for using internally consistent VDF such as fluid queue based VDF
run iteratively through VDF for estimating travel time for the entire analysis period
--> generate VDF based estimated travel time in path_flow.csv
--> generate OD-level VDF based travel time in accessibility.csv
To get more detail our modeling approach please visit DTALite traffic flow modeling white paper
To capture queue formation, spillback, and dissipation through simplified traffic flow models on both meso and microscopic scales, DLSim uses DTALite's computational framework to utilize a minimal set of traffic flow model parameters, such as outflow, inflow capacity, and storage capacity constraints, in its cross-resolution simulation engine, which is illustrated in Fig. 8.
Fig.8: Modeling traffic dynamics
The most simple queue model implemented in DLSIM is the point queue model. By imposing a single outflow capacity constraint on each link, a point queue model aims to capture the effect of traffic congestion at major bottlenecks, although it does not take into account the queue spillback and the resulting delay due to storage capacity. Using a point queue model in the first few iterations and then applying a simplified kinematic wave model in the late assignment process, one can avoid unrealistic and unnecessary gridlock in the initial assignment process, and further allow agents to learn travel times from previous iterations and switch routes to achieve a smooth and close-to-reality traffic pattern.
Fig.9: Point queue model
Figure 9 represents a point queue as a vertical queue with a stack of vehicles, where some of the vehicles are mapped or “rotated” from the physical link (shaded) to the vertical stack queue. The other vehicles on the physical link (not shaded) correspond to the vehicles that move at free-flow speed. In this case, the length of the queue segment in this point-queue model is zero and the link has unlimited storage capacity.
The realism of a point queue model can be enhanced by adding spatial storage capacity constraints, so that the resulting spatial queue model can capture queue spillbacks. This is accomplished in DLSIM using link-specific jam densities, identifying how many vehicles can be stored on a link when no traffic is moving. Furthermore, by explicitly using the cumulative arrival and departure curves to track kinematic waves, Newell’s flow model provides an effective means to realistically represent traffic congestion propagation and capture shockwaves as the result of bottleneck capacities.
In DLSIM, we use cellular-automata CA(M) model where vehicles can move to the next discretized cell where if this cell has been empty for ω − 1 intervals (ω: backward wave). Otherwise, vehicles have to wait in their current cell.
Along this line, we present the conceptual design in Figure 10 where green arrow is shortest path arc, blue arrow is possible delayed arc because of backward wave, bottleneck, etc., orange arrow is excessing arc the shortest movement. In cell-based CA(M) approach, we use 7m for each space difference of consecutive cells and 0.25 s for time axis.
Fig. 10: Space-time trajectory and vehicular event list
For further information please visit:
Research Report by Carlos F.Daganzo, "In Traffic Flow, Cellular Automata = Kinematic Waves"
*** pls refer to Cafer's dissertation
*** pls refer to Part B paper.
insider simulator for VMS and real time information users, consider additional simulated travel time from the queueing
for each simulation time interval
end for
--> generates the dynamic link_performance file
--> generate averaged simulated travel time in path_flow.csv
--> generate the experienced path-level travel time for individual vehicles in agent.csv
--> generate the experienced link-level time sequences for individual vehicles in trajectory.csv
macroscopic process
use modified travel time in VDF based routing
mesoscopic process
reduced time-dependent outflow capacity in the queue-based simulation
reads static and dynamic files static MOE is displayed as text through the configuration tool, link volume and speed visualized at time = 0 dynamic MOE is displayed through dynamic MOE dialog, link volume, speed and density, visualized at time > 0 during 24-hour horizon
Macroscopic-level data exchanges link-based static signal timing data based cycle time, green time for each analysis period link-based static capacity, perceived travel time for routing through LR prices and credit on entire analysis period route suggestions for each OD pair for different agent types (e.g. sov, hov, trucks)
Mesoscopic-level data exchanges:
link-based dynamic capacity, signal timing data for each second/min
route suggestions for each OD pair, information classes, individual vehicles