This document describes approaches for microscopic and macroscopic railway simulations with incomplete data, in particular by comparing simulations where freight trains adhere to the timetable with the case when they are allowed to depart before the scheduled departure time when possible.

The report describes the selected case study along with the setup and simulation process for microscopic and macroscopic simulation for the selected tools RailSys and PROTON. Differences between the tools include how dwell time variation distributions are used and the fact that RailSys requires a workaround to simulate early train departures since it does not accept negative values as initial delays. The process of creating input delay distributions from empirical data is also described and includes downscaling of empirical data to estimate the level of primary delays, and to remove the influence of secondary delays.

Comparing the simulation results it is shown that both for RailSys and PROTON, allowing early freight train departures results in the punctuality being closer to empirical data. Based on the simulations run so far, PROTON has produced results slightly more in line with the empirical data, but this may be explained by differences in the modelling approaches. Further investigation is necessary before final conclusions can be drawn in a comparison between RailSys and PROTON, but the simulation methods seem to correspond satisfactorily.