Simulation-based engineering has become a cornerstone when designing new systems or products. This approach is coupled with an optimization phase in which the result of the engineering is tailored to optimize different aspects of the design (sizing components, minimizing cost, etc.). In the case of complex systems, such as cyber-physical systems, their modeling and simulation calls for multi-domains expertise. As a consequence, a simulation-based approach that integrates different simulators is needed, this is called co-simulation. This article presents a framework situated in the co-simulation basis for implementing optimization, the objective is to sustain the robustness of the multidisciplinary approach and benefit from the large variety of simulation-based optimization algorithms. This framework will be proposed based on existing applications in engineering as well as some study cases to show the benefits of reusing architectural patterns. The implementation allows users to change optimization methods and/or co-simulation elements with the modification of fewer than 10 lines of code.