Cyclodextrins are widely used for the solubilisation of poorly soluble drugs in the formulations. However, current cyclodextrin formulation development strongly depends on trial-and-error in the laboratory, which is time-consuming and high cost. The aim of this research was to compare three modeling approaches (Docking, molecular dynamics (MD), and quantum mechanics (QM)) for cyclodextrin/drug complexation. Ibuprofen was used as a model drug. Binding free energy from three simulation methods was calculated as an important parameter to compare with the experimental results. Docking results from AutoDock Vina program showed that the scoring of complexation capability between ibuprofen and cyclodextrins is alpha (α), gamma (γ), beta (β), and HP-beta-cyclodextrins, which indicated similar ranking with the results from phase, solubility diagram experiments. MD simulation indicated that ibuprofen could form the stable complexes with β-, γ-, and HP-β-cyclodextrins, but not for alpha cyclodextrin. Binding free energies from the MD simulation for β-, γ-, and HP-β-cyclodextrins were -3.67, -0.67, and -3.87 kcal/mol, individually. The enthalpies of QM simulation for β-, γ-, and HP-β-cyclodextrins were -17.22, -14.75, and -20.28 kcal/mol, respectively. Results from all three modeling approaches showed similar ranking between ibuprofen and four cyclodextrin molecules as the experimental data. However, MD simulation with entropy calculation had the closest value to experimental data for β and HP-beta-cyclodextrins. Thus, MD simulation with MM-PBSA method may be fit to in silico screen for cyclodextrin formulations. © 2015 Runmiao Wang et al.