Trafikverket, the Swedish Transport Administration, has been commissioned by the Swedishgovernment to make plans for new railway mainlines between the country’s three largest cities:Stockholm and Gothenburg/Malm¨o. The planned operating speed of the new mainlines isbetween 250–320 km/h. Train traffic at such high speeds across areas with soft soils may lead tolarge amplifications of the train induced ground vibrations, a phenomenon known as the groundvibration boom. The first observation of this phenomenon in practice occurred in Sweden in 1997,at the site Ledsg˚ard on the West Coast Line. This observation led to extensive investigationsand research. One of the research projects consisted of the development of the numerical calculationsoftware VibTrain by the Norwegian Geotechnical Institute, through which high speedtrain induced ground vibrations could be analyzed.The aim of this thesis has been to investigate whether VibTrain also could be used in planningworks for the new mainlines, specifically with respect to (i) evaluating the risk of the groundvibration boom and (ii) designing soil improvement measures. The investigation was conductedthrough calculations using the VibTrain software in three separate parts: (i) a verification of Vib-Train’s function on a modern computer by comparison with previous analyses for the Ledsg˚ardcase, (ii) a parametric study of the calculation model used in VibTrain and (iii) a comparison ofresults from VibTrain with results from the calculation software ωFE-N (¨Ulker-Kaustell 2016)for the location J¨arnasl¨atten in Ostl¨anken, a section of the new mainlines.The main findings from these calculations were:• The use of VibTrain on a modern computer worked well, and the verification of the program’sfunction through comparison with the Ledsg˚ard case yielded consistent results.• Modeling of soil improvement measures in the form of lime cement columns in VibTrain,using the method of equivalent modulus, gave results which principally agreed with theexpected behavior.• The comparison with results from the calculation software ωFE-N showed that the twoprograms agreed relatively well for a case with unimproved subgrade. In the analyzedcases with improved subgrade there were larger discrepancies, due to different modelingstrategies being used in the two programs.Based on the performed calculations, it is evident that VibTrain is best suited for initial assessmentsof the risk of the ground vibration boom. However, the calculation model is not sufficientlydetailed to be used in the design of soil improvement measures for the subgrade.