The aim of this study is to perform a statistical analysis on a pharmaceutical roller compaction process using an industrial-scale roller compactor "Freund TF-MINI model". The process was modelled using response surface methodology (RSM) to better understand and control the process in order to produce ribbons and granules with optimised quality. The significant process parameters were determined to be (i) the screw speed to roll speed ratio and (ii) the roll pressure. The roll speed was kept constant and the roll gap was uncontrolled. The quality attributes of interest are: ribbon density, granule size (D10, D50, D90), amount of fines (granule size < 157 mu m), and tablet hardness. Microcrystalline cellulose (MCC) PH 102 was used as a model material. Design-Expert V9 was utilised to establish the design matrix and to analyse the experimental data. The relationships between the process parameters and the resultant ribbon/granule/tablet characteristics were established. This was followed by an optimisation of the process parameters to obtain the target responses. The results confirmed the attainment of significant models with satisfactory accurate measures. The optimisation allowed for the determination of the process parameters required to produce the best quality tablets. (C) 2016 Elsevier B.V. All rights reserved.