Statistical Comparison of Sumo's Count-Based and Countless Traffic Generation Tools

Simulation of Urban Mobility (SUMO) [1] is a widely used software for traffic simulation. Traffic can be created in SUMO manually or with the help of tools. Basically, the tools can be divided into count-based and countless tools. While countbased tools require data on the traffic they reproduce, countless tools do not. In this work, tools from both approaches are compared quantitatively. Design of experiments is used for this purpose. On the one hand, the count-based tools routeSampler and od2trips were used and on the other hand the countless tools osmWebWizard and randomTrips. The target variables in the design of experiments include travel time, energy consumption of a vehicle on a defined route, simulation time, and the effort required to create the scenario. The results show that with an appropriate choice of parameters, scenarios can also be created using countless approaches, delivering target values comparable to count-based approaches in terms of consumption and travel time. A major advantage of countless approaches is their minimal setup effort for generating traffic demand.