The modelling part of this paper presents two macroscopic motorway network simulation programs METANET and MCONTRM: METANET is based on a purely macroscopic modelling approach. The considered networks may contain several origins, destinations and alternative routes. The impact of traffic control measures such as collective and/or individual route guidance and ramp metering can be simulated in real time. METANET has been validated with real data from the A6a and A6b motorway network (south of Paris). Validation results are presented which show the good qualitative and quantitative compliance with the measured data. MCONTRM assigns traffic to a network of roads in which a diversion strategy is in operation and predicts the macroscopic traffic behaviour. Origins and destinations may be distributed throughout the network, and any number of alternative routes may be allowed. MCONTRM can predict the traffic response to various control strategies, and thus be used in aid-to-decision control systems. MCONTRM was validated also with data from the A6a/b network. A good correspondence between predicted and observed values can be seen. The control part of the paper outlines two network control approaches: a) Aid-to-decision approach using OPERA which combines on-line simulation of control actions and a knowledge based system is outlined. b) The Automatic control approach. The automatic control approach presented is based on automatic control theory. In this paper the considerations are focused on traffic control via Variable Message Signs. The control objectives and the performance criteria which can be used are discussed. An elementary network is presented for which a basic control structure is given. Some preliminary simulation results are presented which are obtained with simple control schemes.
Samenvatting