Population balance (PB) models have become the most widely used tool for dynamic modeling of particulate processes. The application of PB models for prediction purposes is attracting significant interest. A parameter estimation and a quantitative validation of PB models should be carried out before the model can be applied for prediction. PB models are large-scale and nonlinear in the parameters. Moreover, the availability of measurements is typically limited, especially at industrial level, which makes the parameters poorly identifiable from experimental data. This paper shows how a systematic method for analyzing parameter sensitivity and collinearity among parameters, provides a subset of parameters that can easily be identified from the available data. A compartmental PB model of an industrial hydrometallurgical leaching plant is developed. Parameter identifiability of the model parameters is analyzed, and experimental data from the industrial plant are used to identify the corresponding subset of parameters and to verify some of the main assumptions of the model. (c) 2005 Elsevier Ltd. All rights reserved.