Multiobjective optimization is one of the key challenges in engineering design process. Due to the non-uniqueness of the solution in this context, a set of evenly distributed solutions is particularly important for the designer. Directed Search Domain (DSD) method is proved to be efficient enough to tackle such a problem. In this paper, we introduce two main modifications of the DSD to make the algorithm simpler for application. They are related to the conduct of the search domain and reformulation of the appropriate single objective optimization problem. The proposed modifications increase the efficiency of the method in computational time with lower number of objective evaluations. A set of test cases demonstrates the capabilities of the new approach.
展开▼
