By its nature a simple kinetic flotation model is limited because simulation of the flotation process is complex and therefore requires the inclusion of a vast number of variables that affect the flotation system. The majority of these variables are not incorporated into a kinetic model. In an attempt to circumvent this, EMC has for some years been developing an empirically driven float-ability model based on Kelsall's classical two-component equation for which measurements required to calibrate and populate the model can be derived from bench-scale experimental data. The advantage of a kinetic model is that it is quick and inexpensive to apply and the challenge is to accurately model changes in particle properties which result in a change of ore floatability distribution. This paper outlines the progress made in this regard. A case study on a nickel ore is presented in which the effect of frother and plant operation (as changes in cell air rate and mass pull) on froth selectivity and plant performance has been modelled. Modelling was performed using bench-scale generated flotation kinetics derived from tests on various streams with different particle properties within the circuit. The values of flotation kinetics associated with this behaviour identified how frother dosage limited circuit operability and overall performance. This understanding highlighted the steps necessary to optimise performance and was instrumental in implementing the two-concentrate process. The case study illustrates how the use and scope of a simple empirical model may be extended in flotation circuit optimisation methodology. (C) 2008 Elsevier Ltd. All rights reserved.