An optimization tool for offshore bottoming cycle and heat recovery steam generator (HRSG) design has previously been developed. The tool is based on empirical correlations to obtain hydraulic and thermal quantities for the HRSG. However, as these correlations are based on experiments with typical onshore designs, they may not be valid for the compact designs encountered in offshore HRSGs.In order to extend the validity range of the optimization tool, this work presents a numerical model able to predict heat transfer and pressure loss in finned tube bundles by means of Computational Fluid Dynamics (CFD), utilizing a periodic domain to reduce computational costs. Both steady-state and transient models were applied, using the Spalart-Allmaras turbulence model, and their performance compared. To validate the model, results were compared with available experimental data, and then the model’s performance was compared with a selected empirical correlation.Three different fin-tube geometries were investigated (two serrated and solid) with varying tube layout angles. A parameterized grid generation tool was developed and used to generate grids for the selected geometries. The CFD results were found to be within 20 % of the experimental data, and were in most cases more accurate than the empirical correlation. The steady-state simulations did, however, not converge for the geometry with the largest layout angle. The steady-state framework should therefore be applied only to compact tube layouts. The transient simulations, though being computationally more intensive, are also able to model large layout angles.