Aurora vs TransModeler

[Andy]

Hi all,

I have uploaded:

1. A ppt file (TransModeler_meso.ppt) containing a summary of the
mesoscopic model in TransModeler (including the model of signal
controller);

2. A spreadsheet (contour_plots.xls) containing the numerical results
of a working example which compares the properties of Aurora (CTM) and
TransModeler. I used Vera's network:

i) 6 links in series which are all 1 mile long;
ii) only one O-D pair connected with 1 route;
iii) capacity of each link is 6000 vph;
iv) demand is 4800 vph throughout the simulation horizon;
v) the simulation horizon is 1.5 hr (i.e. 90 min)
vi) an incident happened during time 30-min and 45-min in which the
capacity of the fifth link drops to 1500 vph.

The results show that Aurora-CTM is able to generate a plausible
contour plots of density, flow, and speed and i can fully understand
the CTM simulation results of the scenario. However, I can't
understand the numbers generated by TransModeler. We can discuss it
further when we meet..

Please feel free to comment.

Andy

[Andy]

Some further note: as Gabriel pointed out today, TransModeler seems to
be unable to capture the "capacity" effect. If you read the number on
the uploaded spreadsheet, TransModeler would give some flow exceed
capacity (6000 vph) at some time intervals ...

Andy