The Odenplan Station in the Stockholm City Link project is located directly under theexisting subway station with a minimum rock cover of approximately 7 m. The span of theOdenplan Station is 25 m. When the Odenplan station was excavated, deformations largerthan anticipated in previous calculations were observed. The main reason for this discrepancyis believed to be high horizontal in situ and/or persistent fractures in the area parallel withthe excavation.The excavation of the Odenplan station constitutes a case study with well-documented rockmass characteristics, excavation sequences, support sequences and measured deformations.It therefore provides a unique project to analyse. Based on this case study, the main objectivesof this study was to (1) suggest a general technique for numerical back analyses based onmeasured deformations under geological conditions as those present in the Stockholm area,and (2) analyse the range of possible in situ stresses at Odenplan Station in the City Linkproject.The complex three-dimensional 3DEC model lead to rather long computational times. Ageneral technique for a real back analysis was therefore not possible to obtain in this study,since it would require unrealistically long computational time. Instead, a sensitivity study onthe in situ stresses and different types of continuum and discontinuum approaches wereperformed. The results from the analyses gave valuable insight into which types of modelsthat were able to properly recreate the observed ground behavior at Odenplan. The resultsfrom the study clearly showed that a continuum approach does not capture the true behaviorof the blocky rock mass at Odenplan. It is therefore recommended that discontinuumapproaches are used in the future when structurally controlled block movements could beexpected. However, the exact cause behind the discrepancy between the continuum and thediscontinuum model has not been determined in this report.The performed sensitivity analysis suggests that the in situ stresses are high and in the rangeof those previously measured. Maximum horizontal in situ stresses are probably in the rangeof 5.7-7.2 MPa at a depth of 0-20 m, which agrees well with the results of previous rockstress measurements, which were, on average, 8.1 MPa at a depth of 30 m.The encouraging results from this project indicate that further stress sensitivity analysesshould be performed to have a better estimation of the range of in situ stresses in theStockholm area by fitting displacements obtained from modeling to those measured in situ in different projects. These analyses are important in order to understand our present abilityto perform numerical simulations reflecting true rock mass behavior.