Dear all,
I am working with a real-life network in SUMO and could need your advice for calibrating the simulation.
I have an OD-matrix and a network given. Further, I have traffic count data from edge counts in the city center (but not from the rest of the region).
So far, I have created TAZ's according to my OD matrix and generated trips with od2trips. This comes with the benefit that origins and destinations are spread over the whole TAZ.
Now, I would like to use routeSampler.py to calibrate the traffic according to the edge count data. My idea so far was to:
- Use duarouter to convert trips to routes (since origins and destinations spread over all edges this would basically create a route for each trip)
- Use this set of routes and my edge counts as input for routeSampler.py
However, I see two major problems with this approach.
- Duarouter creates all routes based on free-flow speed and does not include alternative routes which would only be used during congestion.
- All traffic count data is from the city center, therefore, routeSampler would not use any route that does not pass an edge with traffic counts (e.g. traffic that bypasses the city center).
The first problem could probably be solved by using duaIterate.py, creating a set of routes in the dynamic user equilibrium.
But the second point seems to be tricky. In the
documentary, it is also briefly mentioned that OD files can be used, but only edge-based. Could I work with this option to make sure the routes still fit the OD more or less?
Are my thoughts correct? Would it be better to use another tool to try for example to scale the OD as a whole to fit the traffic counts?
Thanks a lot for your help in advance!
Best regards,
Joel
