<p>Financial markets are considered as a main base for economic growth. Due to the rapid development of these markets, the portfolio optimization problem has become one of the most complex problems in finance. This paper addresses the multi-objective portfolio optimization problem (MOPOP) with three different objectives, which maximize the return, minimize the risk, and maximize the entropy to generate a well-diversified portfolio. We adapt the original manta ray foraging optimization (MOMRFO) to handle the MOPOP issue. Our adapted MOMFRO is called MOMFRO/ED as it uses an epsilon dominance relationship to store the non-dominated solutions in an external archive. Furthermore, the external archive population is controlled using crowding distance to limit the archive size and avoid increasing the complexity of the MOMRFO/ED algorithm. The best solutions (leaders) are selected from the external archive population to improve the solution’s quality and accelerate the convergence computational. Several experiments are conducted on real-data from the major financial markets during two different periods (before and within the COVID-19 pandemic period) to compare our algorithm with three relevant state-of-art algorithms. The statistical analysis of the obtained comparative results shows the merits and the outperformance of our MOMRFO/ED algorithm in terms of IGD, HV, SP, SR, and Jensen Index (JI)metrics. In addition, our results indicate that model with Shannon’s entropy outperform those with Minkowski and Yager’s entropy. This superiority is attributed to the enhanced ability of Shannon’s entropy to efficiently reallocate assets in response to market changes.</p>