1 Department of Civil Engineering, University of British Columbia, Vancouver, Canada. Electronic address: l.abkar@ubc.ca.
2 Department of Chemical and Materials Engineering, Concordia University, Montreal, Canada.
3 OASEN Drink Water, Gouda, Netherlands. Electronic address: morez.jafari@oasen.nl.
4 Department of Civil Engineering, University of British Columbia, Vancouver, Canada. Electronic address: sara.beck@ubc.ca.
5 Department of Civil and Environmental Engineering, University of California Merced, Merced, United States of America. Electronic address: aghassemi2@ucmerced.edu.
6 Department of Environmental Biotechnology, Delft University of Technology, Delft, Netherlands. Electronic address: M.C.M.vanLoosdrecht@tudelft.nl.

Reverse osmosis (RO) systems offer a viable solution for treating brackish water (BW), a common but underutilized water resource. However, the energy-intensive nature of brackish water reverse osmosis (BWRO) systems poses affordability challenges to water supply, necessitating a focus on minimizing their energy consumption to support SDG6's goal of providing safe and affordable drinking water for all. This study addresses the critical need to minimize the specific energy consumption (SEC) of a typical BWRO system, defined as the energy consumed per unit of water recovered, mathematically and experimentally. Empirical models were developed proving there is a global minimum SEC while adjusting the operating conditions. Furthermore, we identified the key operating factors influencing SEC and their priority levels, along with their interactive effects. Notably, no prior study has discussed the significance and interaction of these operating factors (e.g., feed water salinity, temperature, pressure, flowrate and membrane permeability) on SEC of a BWRO system. Employing a full factorial experimental design with mixed levels of operating parameters, the study developed regression models that elucidate the mechanistic interaction between these parameters and system performance. Moreover, the models were validated experimentally, with a new dataset demonstrating their accuracy and reliability. ANOVA statistical analysis identified feed salinity, pressure, flow rate, feed flow rate×pressure, salinity×pressure, and temperature as influential operating parameters in reducing SEC, in descending order of importance. Operating within the determined optimum range resulted in a 36 % decrease in SEC and a more than fourfold increase in water recovery. The study's systematic approach and findings can be extrapolated to optimize the performance of other desalination technologies and diverse feed water types, contributing significantly to global water sustainability efforts.