Transportation forecasting
Traffic model is a technical term transport planning and deals with the mapping of transport processes in models. Depending on the considered level of detail, there are microscopic, mesoscopic and macroscopic traffic models that consider the traffic demand and traffic flow.
Traffic flow model
These models are used to make statements about the flow of traffic and the traffic density on a given infrastructure. They are usually limited to the ( motorized, rare non-motorized ) traffic. When the traffic flow models microscopic and macroscopic models are used.
The microscopic models form individual driver / vehicle units from, each with their individual characteristics. Due to the associated computational complexity, these microscopic models are particularly suitable for small study areas. They find application in traffic planning for the simulation of the impact of individual infrastructure projects as well as adaptive cruise control to calculate a safe response to changes in speed of the preceding vehicle application. Known representatives of the microscopic models are the Cell Transmission Model ( CTM), the Nagel-Schreckenberg model or the car-following model by Rainer Wiedemann, which is used by its use in VISSIM in planning practice with the most.
The macroscopic models form the traffic from only in its mass, individual driver / vehicle units are not considered. These models allow statements about the flow of traffic and the traffic density. Such statements can be interpreted with a fundamental diagram. Thus, the traffic status ( ie, whether the traffic is flowing freely or is stowed ) are described. Typical applications for these models are for example the congestion forecast in major networks such as the German motorway network. These models, for example, the Section Model Based heard.
Travel demand model
These models usually work on a macroscopic level.
For necessary is the dividing of the planning area in equivalent cells traffic / traffic areas. Their size, homogeneity and availability of socio-demographic data affects the accuracy of the later model results. Among themselves, the traffic cells are connected by transport lines. Cells traffic and traffic lines together form the network model.
Within the transport model four calculation steps for determining the traffic demand can be carried out:
Traffic generation
The degree of traffic generation a traffic cell determines the transport planners about the existence of a cell function. Depending on whether a cell is used as a residence or as a workplace, different traffic volumes are generated. This data can be extracted or calculated from the statistics. Common calculation models for this are the parameter model, the regression analysis or the improvement factor model. The result is information about the source or destination traffic.
Traffic distribution ( transport destination choice)
By calculating the traffic generation remains unclear which other traffic cells to spread the traffic.
There are different destination choice models. For small study areas and areas where spatial resistors not play a major role, the so-called random model is applied. Here, the calculated in the traffic generation source and target traffic to be distributed in proportion to the total volume of traffic on the traffic cells.
One of the first models, which also took into account distances, was the so-called gravity model, which was applicable from 1889 in Lill'schen travel law. In this calculation, the traffic distribution is subject to the assumption that a traffic cell as a gravitational point behave, ie a cell gets more attraction, the more mass it has. With increasing distance of the attraction of the cell is increasingly smaller. This gravity model comes from the mechanics and gives the traffic distribution within the planning area relatively accurately again.
In the logit model, it is assumed the one-touch consists of a deterministic and a stochastic part. As an evaluation function for the traffic a natural exponential function is used.
In the basic model of target selection evaluation functions are multiplied as the above factors with the source and target -side, as well as Modifeinen. These factors are determined so that the traffic distribution corresponds to the previously determined individual traffic. with the boundary conditions:
, with
, with
, with
A possible calculation of these factors is the so-called Furness algorithm.
The problem with the above models is, among other things, that they are very sensitive to small changes in closely spaced Aufwänden. This disadvantage can be compensated from the so-called EVA -1 function, so that only when greater differences, the transport current distribution differs significantly.
The result is (also OD matrix ( origin -destination ) ) set out in a square traffic stream or source-destination matrix.
Choice of transport ( modal split )
The choice of transport is the distribution of traffic on individual ( MIV = motorized individual, NIV = non- motorized private ) and public transport (PT ) - the so-called modal split - determined. It must be considered also the pedestrian and bicycle traffic (NIV ) to allow the subsequent calibration of the data collected. In order to obtain realistic values for the calculation, the correct choice of parameters have to be taken. In the choice of transport mode, a distinction between three use cases:
TripAdvisor end / Trip Interchange
The previous two steps can be applied depending both in this order (so-called trip - interchange model (TIM ) ) and vice versa ( so-called trip -end model ( TEM) ). In the TIM method first the distribution of traffic volumes is made between transport cells, and later distributed to the individual modes. In the TEM method, the entire traffic is first distributed to the various modes and distributed later to the individual transport relations. Depending on the chosen distribution method, these methods may lead to different results.
Traffic assignment (Traffic routing )
The values obtained in the calculation method mentioned above can be represented in a matrix. In the traffic assignment determines which route the traffic to selected from the source to the destination. The traffic planner can choose between four models of computation: