This study reports a study on the process modelling of ultrafiltration (UF) units with a focus on the development of backwashing models based on properly designed experiments under pilot scale conditions. For the modelling, series of experiments were designed using a star- 23 (fractional) factorial design methods and performed with the so-called SMART-XIGA pilot plant provided by Norit Membrane Technology, a key player in the membrane manufacturing. The influential factors on both the Hydraulic Backwash (HB) and Chemically Enhanced Backwash (CEB) are considered as variables in the resulting second-order regression models. The development and cross-validation of the models is solely based on the results obtained from the properly designed experiments. Explanation of the interactions among the considered factors was investigated via graphical and eigenvalue analysis using Response Surface Methodology (RSM) approach of empirical modelling. The results show that, a change in transmembrane pressure in HB is largely dependent on the backwash time and backwash frequency with backwash flux having no effect. Reversibility of the fouled layer in CEB is greatly dependent on the coagulant concentration dosing and the filtration flux and not so much on the filtration time. Also, the cross-validation and the ANOVA analysis carried out show that the models are valid in the region of our nominal working points.