This paper presents a methodology for the prediction of the performance of sea water reverse osmosis units acquired by salient mathematical correlations which involve the important aspects of membrane input and output variables. Feed water conditions pertinent to flow rate, TDS (total dissolved solids), pressure, pH and temperature are correlated with recovery ratio, permeate TDS and power consumption via regression analysis. Satisfactory results are obtained by comparing the resulting correlations against the software ROSA predictions. The mathematical correlations and regression assessment led to the derivation of simple polynomial expressions which were used for sensitivity analysis and may be useful for the design, monitoring and control of this important desalination process. (C) 2010 Elsevier B.V. All rights reserved.