next up previous
Next: Summary Up: Cellular automata models for Previous: Limits of the CA


A simulation of all of Switzerland

One of the current main goals in our group is a simulation of ``all of Switzerland''. By this we mean a microscopic 24h simulation of a typical workday of all traffic in Switzerland. Fig. 4 contains an early result of this.

The network that is used was originally developed for the Swiss regional planning authority (Bundesamt für Raumentwicklung), and has since been modified by Vrtic at the IVT and by us. The network has the fairly typical number of 10572 nodes and 28622 links. Also fairly typical, the major attributes on these links are type, length, speed, and capacity.

Demand is obtained from a 24-hour origin-destination matrix with 3066 zones, also from the Bundesamt für Raumentwicklung. This matrix is converted to 24 separate hourly matrices by a several-step heuristic. In the long run, it is intended to move to activity-based demand generation. Then, as explained above one would start from a synthetic population, and for each population member, one would generate the chain of activities for the whole 24-hour period.

Routes are obtained via iterations between simulation and time-dependent fastest path routing. The simulation behind Fig. 4 is the queue simulation as described in Sec. 5.

Figure 4: Most of Switzerland at 8am, simulation result. The graphics shows individual vehicles, but they are so small that they cannot be distinguished. Areas in dark contain traffic jams.
\includegraphics[width=\hsize]{ch-8am-tifgz.eps}


next up previous
Next: Summary Up: Cellular automata models for Previous: Limits of the CA
Kai Nagel 2002-05-31