13.4.6 Network Loading
The next step in coding a Vissim model is adding vehicles, bicycles, and pedestrians to the network. For the purposes of this section, the word “vehicles” includes vehicles, bicycles, and pedestrians.
Vissim uses the European fleet as a default when beginning a new network. Change this setting to the American fleet to reflect longer vehicle lengths in America.
2D/3D models are the basis for the vehicle mix in the Vissim model. Each 2D/3D model is assigned a *.v3d model with a length of the vehicle that is simulated in the network. The distribution of the 2D/3D models is assigned using the 2D/3D model distributions, which establish the mix of 2D/3D models that appear in the network. The 2D/3D model distribution is assigned to the Vehicle Types. This provides a variety of vehicle lengths for each Vehicle Type used. For example, there can be several 2D/3D models that represent different models of cars. The 2D/3D distribution determines the ratio of each model of car expected to be seen in the network. The Vehicle Types for cars is assigned to this 2D/3D distribution so that the cars in the network show the correct ratio of each car model.
Vehicle Types are made of 2D/3D models. These aid in setting the variety of vehicles in the network. Vehicle Types are assigned to a specific Vehicle Composition that is used in a Vehicle Input to appear in the network.
Vehicle Compositions describe the type of vehicles simulated and are made of Vehicle Types. The Vehicle Composition is assigned to each individual vehicle input and determines the flow of different vehicle types. Vehicle Compositions typically have one or more Vehicle Type assigned to it. The Relative Flow determines what portion of the vehicle input volume is simulated as each Vehicle Type in the Vehicle Composition. For example, if a Vehicle Composition has three Vehicle Types with a total Relative Flow of 1 and a Relative Flow for Vehicle Type A of 0.25, then 25 percent of the vehicle input volume associated with that Vehicle Composition is simulated as Vehicle Type A.
Vehicle Classes are used to group Vehicle Types. Vehicle Classes typically contain any number of Vehicle Types, with the option to contain all Vehicle Types. Vehicle Classes provide the basis for speed data, evaluations, path selection behavior, and other network objects. For example, it is possible to restrict a Vehicle Class from entering a specific lane in a network link. The Vehicle Class does not determine the mix of vehicles in the network; this is determined by the Vehicle Composition.
Vehicle Inputs determine the volume of vehicles entering the network. These are assigned to specific links in units of vehicles per hour. Vehicle Inputs can either be stochastic (varies according to random functions based on the seed number for a given run) or exact (exact number of vehicles simulated as indicated by the flow rate). Microsimulation projects typically involve stochastic modeling of vehicle inputs. shows the relationship between the different network loading characteristics. Vehicle characteristics are shown in .
Characteristic | Summary |
|---|---|
Vehicle Type | Made of 2D/3D vehicle model distribution. |
Vehicle Composition | Made up of Vehicle Types. Vehicle composition is used to assign the percentage makeup of the traffic entering the network via Vehicle Inputs. |
Vehicle Class | Used to group Vehicle Types; can contain any number (or all) Vehicle Types |
Vehicle Input | Volume of vehicles entering the network; assigned to links in units of vehicles per hour. |
Vehicle Route | Determines where vehicles travel in a network. Most common route types are Static and Partial. |
Figure 13-5: Relationship between the different network loading characteristics
Vehicle Routes are coded using Vehicle Routing Decisions and determine where the vehicles travel within the network. Relative Flows determine how many vehicles are assigned to each route based on the total number of vehicles that arrive at the Routing Decision during the time interval. There are many different types of Vehicle Routes, but the two most common are Static and Partial.
- Static Routes are used for base routing of network volumes. Static Routes are chosen to create relay routes that provide a new Routing Decision at each decision point. Static Routes can also be chosen to create end-to-end routes that account for travel patterns in the Vissim network and are often based on trip tables. Vehicle Routes default to the path of least resistance, but routes can be manually adjusted by the analyst. Static routes are typically used when no parallel routes exist in the model.
- Partial Routes are used for a secondary routing system of network volumes. Partial Routes can be used to control the Relative Flow of vehicles that are able to use multiple paths.
It is important to establish a consistent numbering scheme when coding Vehicle Inputs and Vehicle Routes. These numbering schemes are used when assigning inputs and routes at intersections.
DTA can be used in circumstances where parallel routes exist (i.e., multiple paths to and from an OD pair). Setting up a model with dynamic traffic assignment demand determining an OD matrix, coding nodes at the network corners, inserting nodes at diverging internal paths, and coding zones that are used as the origins and destinations. Using DTA can be a difficult and time-consuming process. It is recommended that DTA only be used if it is critical and cannot be substituted by static or partial routing.