A mathematical model for making quantitative and qualitative predictions regarding the performance of water-cooled PV/T systems is presented. The hypothesis of the research is that design parameters affecting photovoltaic/thermal collectors' feasibility have to be optimized simultaneously in order to choose among a set of conflicting solutions the optimal design which fits a given application. The collectors design and performance are studied simultaneously using an elitist multi-objective evolutionary algorithm Non-Dominated Sorting Genetic Algorithm-II (NSGA-II). NSGA-II derives the Pareto optimal set, which illustrates the trade-off between solutions. (C) 2014 Elsevier Ltd. All rights reserved.