In order to use the potential of drop-population balance simulations for the design of Kuhni pilot-plant columns, a quantitative understanding of the effect of compartment geometry on drop residence-time distribution is necessary since this is crucial for the accuracy of the simulation. This work presents an experimental method based on single-drop experiments. With this method, the effect of compartment geometry and impeller speed on the drops' residence-time distribution can be quantified systematically. Drop residence times are evaluated not only for the whole compartment but also for different compartment zones, namely the lower zone between bottom stator and impeller, the impeller zone, and the upper zone between impeller and upper stator. It is also shown that by reducing only the height of the lower compartment region, the compartment height can be reduced by 20% without a significant decrease in drop residence time. Based on the experimental findings a stochastic model comprising three sub-models, one for each individual zone (lower toroidal vortex, impeller region, upper toroidal vortex) is developed. In this model, drop movement is described by superimposing drop sedimentation velocity onto the axial velocity of the toroidal vortex, which is a function of the drop's radial position. The model uses stochastic methods to account for stator impact or drops being back-mixed between the individual compartment zones. With this model, it is possible to quantitatively describe not only the mean residence times of drops but also the residence-time distribution for different drop diameters, impeller speeds and open-area fractions of the stator. (C) 2013 Elsevier Ltd. All rights reserved.