In regions with cold climates, a layer of ice often forms on vehicle windshields, whichobstructs the driver’s view. To address this issue, vehicles are equipped with internal defrosters. However, at Scania, the evaluation of defroster design performancecurrently relies on time-consuming and costly physical testing. A more effectiveapproach would be to employ numerical techniques to accurately forecast defrostingpatterns. This would offer valuable insights for analyzing the defroster’s performanceduring the design phase.The objective of this thesis is to develop a methodology using the open-source CFDsoftware OpenFOAM to predict the performance of a vehicle’s defrosting system.This approach presents a quicker and more convenient way to design the systemcompared to conventional testing methods. Experimental results were obtained bymonitoring the defrosting process at regular intervals. However, uncertainties existedregarding boundary and ambient conditions as the experiments were not conductedto validate the CFD results. The temperature profile and mass flow rate at the inlet were unknown. The model’s geometry was pre-processed using ANSA, and thevolume mesh for the truck cabin was generated using the SnappyHexMesh utilityin OpenFOAM. Mesh verification demonstrated good quality, and the realizable k-εturbulence model was utilized. The Grid Convergence Index (GCI) was employedto compare different mesh sizes, ultimately achieving a converged mesh. The RKEmodel was found to be computationally efficient and suitable for defrosting simulations, producing similar results to the k-ω SST turbulence model.A time step study was conducted to determine an efficient time-step. Additionally,a temperature study was performed to address the uncertainty surrounding the inlet temperature. Various design points were examined, involving different heat-uptimes and maximum temperatures. The results indicated that a heat-up time of 600seconds and a maximum temperature of 308 Kelvin yielded similar outcomes to theexperiments. To address uncertainty regarding the inlet mass flow rate, a study wasconducted by varying the mass flow rate. Comparing the results with the experimental data, a mass flow rate of 450 kg/hr provided the most comparable defrostingperformance. The study also investigated the impact of the exterior domain anddetermined that removing it would lead to inaccurate defrosting predictions due to alack of heat transfer. Furthermore, a comparison of OpenFOAM and StarCCM+ forsteady-state solutions demonstrated satisfactory results in terms of turbulent kineticenergy and wall shear stress at the windshield. Attempts to optimize defrosting performance included optimizing the shape of the defroster vents. The effect of rotatingthe vents relative to the windshield surface on defrosting was assessed, but it wasconcluded that the angle had minimal impact on performance or the methodology isnot sensitive enough to differentiate the minor differences.In conclusion, this thesis presents an efficient methodology utilizing OpenFOAM topredict defrosting performance, encompassing complete windshield defrosting timeand ice melting rate. It holds potential for future defroster design processes. Furtherstudies could focus on alternative meshing methods to reduce computational costs.