The objective of this research was to develop a modified genetic algorithm for solving the multi-objective optimization problem focused on Pareto based approach. Multi-Objective Bisexual Reproduction Genetic Algorithm (MOBRGA) is a new GA proposed herein. MOBRGA uses a Darwin’s sexual selection concept, and different types and rates of mutation when creating offspring. When comparing MOBRGA with other algorithms, the results showed that MOBRGA was competitive to others in some problems such as Shaffer’s and Murata’s problem. From experiments, MOBRGA could show the Pareto front. It could find non-dominated solutions to the problems. In the other words, it could solve the multi-objective optimization problems. As an example of applications, MOBRGA was applied for a computer network design. A computer network design is a multi-objective optimization problem, where many objectives are simultaneously considered. When designing computer network using MOBRGA, a set of solutions that met the specification of products was resulted. The solutions found the user’s requirements and products supported with each other. It can be concluded that the proposed algorithm can be effectively applied to real applications.