By maximizing a number of competing goals all at once, multi-objective optimization (MOP) is an indispensable tool for handling difficult issues in engineering, business, and decision-making. Since the decision-maker’s preferences and priorities aren’t taken into consideration by traditional MOP approaches, the enormous solution sets they produce may not be suitable for real-world decision-making. This work introduces a new method for optimizing many objectives simultaneously by using mathematical models that take into account the preferences of decision-makers. To tackle the issue of search strategy parity, a hybrid AI method is suggested that merges Genetic Algorithms (GAs) with Particle Swarm Optimization (PSO). To further account for the ever-changing nature of limitations, an adaptive technique is presented for addressing them, which allows penalty factors to be dynamically adjusted. Tests show that the suggested architecture greatly lessens computing load, makes solutions more relevant to decision-maker objectives, and offers strong handling of complicated constraints in optimization problems. Practical multi-objective optimization issues involving preference-guided solutions and dynamic constraints are greatly improved by this approach.