A Pareto set coordination method for analytical target cascading

In the general analytical target cascading method, a weighted-sum formulation is commonly employed to coordinate the inconsistency between design points and assigned targets at each level while minimizing the cost. The determination of weighting coefficients is problem dependent. Improper selections of the weighting coefficients may result in incorrect solutions. To avoid using the weighting coefficients, a genetic algorithm optimization method is developed for the hierarchical design problem by using the Pareto set coordination method. The Pareto sets are obtained from the optimal solutions at each level while each subsystem chooses one solution based on the detailed information. Instead of setting point-valued targets and weighting coefficients, Pareto sets are computed and updated at multiple levels until targets are satisfied. Therefore, the genetic algorithm optimizer with Pareto set coordination for analytical target cascading can avoid choosing weighting coefficients. By doing so, the proposed method explores completed feasible solutions at each level and improves the convergence process. The results for the proposed method and the weighted-sum analytical target cascading are compared to illustrate the performance of the proposed method.