A traffic simulation system increasing the efficiency of schedule design for public transport systems based on scarce data

This work presents a modular system for the simulation of public transport schedules. Two simulation approaches are implemented and compared regarding speed and compliance of results to reality. One approach serves to analyze various priority queue data structures with respect to their impact on overall execution times of simulation runs. The system is built around a database at first containing sample network and schedule data of the trams in Cologne provided by the Kölner Verkehrsbetriebe (KVB) AG. Other data could be incorporated easily due to the standardized data structure. A simple platform independent graphical user interface (GUI) module enables near-live analysis of simulation results. Comparing the priority structures yields only a very small advantage for a custombuilt structure over a more general priority queue structure, but a much larger time penalty for using an optimized heap structure directly suitable also for sorting and time rollback.