The input data consists of two parts: the street network, and the demand.
The street network that is used was originally developed for the Swiss regional planning authority (Bundesamt für Raumentwicklung), and covered Switzerland. It was extended with the major European transit corridors for a railway-related study (22). The network supposedly contains the status for 1999, but contains at least one major error (a high capacity tunnel in Zürich is missing). Our initial simulations resulted in traffic gridlock in Zürich, which was also reflected in the VISUM assignment displaying V/C ratios significantly above 100%. A manual comparison with a higher resolution network of Zürich led to the conclusion that capacity in Zürich was in general significantly underestimated; in consequence, we manually increased the corresponding road capacity for transit corridors through Zürich in our network. We can only speculate what led to these network errors; Sec. 7 discusses our plans of how to improve the situation.
After our modifications, the network has the fairly typical number of 10564 nodes and 28622 links. Also fairly typical, the major attributes on these links are type, length, speed, and capacity. As pointed out above, this is enough information for the queue simulation.
In order to test our set-up, we generated a set of 50000 trips going to the same destination. Having all trips going to the same destination allows us to check the plausibility of the feedback since all traffic jams on all used routes to the destination should dissolve in parallel. In this scenario, we simulate the traffic resulting from 50000 vehicles which start between 6am and 7am all over Switzerland and which all go to Lugano, which is in the Ticino, the Italian-speaking part of Switzerland south of the Alps. In order for the vehicles to get there, most of them have to cross the Alps. There are however not many ways to do this, resulting in traffic jams, most notably in the corridor leading towards the Gotthard pass. This scenario has some resemblance with real-world vacation traffic in Switzerland.
Our starting point for demand generation for the full Switzerland scenario are 24-hour origin-destination matrices from the Swiss regional planning authority (Bundesamt für Raumentwicklung). Eventually, we intend to move on to activity-based demand generation.
The original 24-hour matrix is converted into 24 one-hour matrixes using a three step heuristic. The first step employs departure time probabilities by population size of origin zone, population size of destination zone and network distance. These are calculated using the 1994 Swiss National Travel Survey (23). The resulting 24 initial matrices are then corrected (calibrated) against available hourly counts using the OD-matrix estimation module of VISUM (24). Hourly counts are available from the counting stations on the national motorway system. Finally, the hourly matrices are rescaled so that the totals over 24 hours match the original 24h matrix.
VISUM assignment of the matrices shows that the patterns of congestion over time are realistic and consistent with the known patterns. The Zürich congestion problem, mentioned above, is contained in the assignment, but did not show up at this higher level view; see Sec. 7 for some discussion of this. A more detailed verification of these results was not possible so far, but is planned.
These hourly matrices are then disaggretated into individual trips. That is, we generate individual trips such that summing up the trips would again result in the given OD matrix. The starting time for each trip is randomly selected between the starting and the ending time of the validity of the OD matrix.
The OD matrices assume traffic analysis zones (TAZs) while in our simulations trips start on links. We convert traffic analysis zones to links by the following heuristic:
This leads to a list of approximately 5 million trips, or about 1 million trips between 6am and 9am. Since the origin-destination matrices are given on an hourly basis, these trips reflect the daily dynamics. Intra-zonal trips are not included in those matrices, as by tradition.