Traffic models are central tools in transport planning. An investigation on this issue was part of a master’s thesis at the Technical University of Munich (TUM) in collaboration with the BERNARD Gruppe. Its aim was to show the extent to which different software tools deliver equivalent results.
For this purpose, a calibrated sub-network of the Munich traffic model was reproduced in detail and an identical planning scenario, the construction of a bypass, was modelled in both systems. The parameters capacity, demand and relative gap (model accuracy) were each varied in 30 steps to reveal differences in traffic load and average speed.
The analysis demonstrates that: Transferring a model to another software is possible in principle, but it requires manual effort due to differing modelling parameters, such as cost logic. At the same time, the sensitivity analyses show a high degree of consistency in the results. Within the set limits of the study, the software tools examined delivered comparable results.
Felix John, BERNARD Gruppe
