This study explores the optimization of a small-sized solar photovoltaic-thermal (PV-T) water collector using the Multi-objective Imperialist Competitive Algorithm (MOICA) and compares its performance with the augmented ε-constraint method (AUGMENCON). The collector, designed to simultaneously generate electricity and heat water, was evaluated based on its average thermal and electrical efficiencies. Key parameters such as mass flow rate, inlet temperature, and inclination angle were optimized using MOICA to maximize both efficiencies. The results were validated against the AUGMENCON method and previously reported genetic algorithm results, showing strong agreement across configurations. Notably, the MOICA method yielded a maximum electrical efficiency of 81.687%, aligning closely with the AUGMENCON findings. This research demonstrates that the dual-function PV-T system, when optimized, can enhance energy capture and efficiency, offering a promising solution for small-scale applications where space is limited. The findings validate the robustness of MOICA in solving complex multi-objective optimization problems, offering reliable Pareto-optimal solutions for real-world energy systems.