Showing papers on "Network traffic simulation published in 1974"

Some possibilities for using traffic control to influence trip distribution and route choice

Abstract: If the generalized cost to a vehicle of travelling along each link in a road network is a suitable known increasing function of the traffic flow on that link, and the number of journeys made per unit time between each point of entry to the network and each point of exit from it is a suitable known decreasing function of the generalized cost to a vehicle of travelling between the two points concerned, or is in accordance with a suitable gravity model, then the techniques of trip distribution and traffic assignment, together with suitable assumptions about drivers' choice of route, enable estimates to be made of the number of journeys between each entry point and each exit point, and the flow of traffic on each link. When all or part of the network is subject to traffic control, the relationships between travel cost and traffic flow on some or all of the links in the network depend on the control parameters, and these can therefore be used to influence the number of journeys made through the network and the routes taken. In this paper, a theoretical framework is developed for such use of traffic control when the method of control is linked traffic signals. The estimated numbers of journeys and traffic flows given by trip distribution and traffic assignment are treated as functions of the traffic control parameters so that these parameters can be chosen so as to improve the traffic pattern with respect to a suitable criterion. An examination is made of the possibility of using trip distribution, traffic assignment and traffic signal calculation techniques to investigate what the results of choosing traffic signal settings in this way might be. A numerical example is given for a simple situation in which a beneficial result could be achieved. /Author/

Smooth: a combined continuous-discrete network simulation language

Abstract: SMOOTH is a packaged network simulation language written in FORTRAN. It is based on the GASP TV and GERT IIIZ simulation languages and is capable of representing and analyzing models which include ...

A generalized street network simulation model

Abstract: This paper describes a microscopic simulation model that reproduces traffic flow on a signalized street network under laboratory conditions. The input format and structure of the program facilitate application to any moderate-sized network. The program is economical to use, achieving a 6.5-to-1 real-time to simulation-time ratio for an 85-link network. The model has undergone a testing and validation procedure in which simulated vehicular travel times have been compared with actual travel times recorded on the test network to evaluate overall model accuracy. An analysis of individual model segments has been conducted to test their sensitivity to changes in key parameters. Results of these tests indicate that the model accurately and realistically simulates traffic flow through the street network.

Stochastic integral equations applied to telecommunications traffic without delay

Abstract: Modern telecommunication techniques case the problem of traffic handling in the framework of fairly general networks, as applied to traffic without delay but with virtually arbitrary service-time distributions. In this paper we use stochastic integral equations to deal with the case involving the most general input process and lost calls. For this purpose, Fortet's equation, unsolved so far in the general case, is solved to analyze the single trunk group model. The stationary case is then treated as a special case. Finally we study networks which satisfy a certain assumption of symmetry. The same general stochastic assumptions are maintained throughout the paper.