Optimization of scale-based product family using multiobjective genetic algorithm

An optimization process using nondominated sorting genetic algorithm-II (NSGA-II) was put forward based on the theoretical model and the optimization model of scale-based product family. Based on the mathematical model of product family design, the Pareto set for the multiobjective optimization problem was obtained using NSGA-II. An approach based on fuzzy set theory was developed to extract one of the Pareto-optimal solutions as the best compromise one. During the first stage, each product in the family was optimized independently with NSGA-II. The design variables that showed small deviations were held constant to form the product platform. In the second stage, the scaling variables of each instance product were developed also using NSGA-II. The performance of instance product was improved on the premise that the design requirements of scale-based product family were satisfied. The efficiency and effectiveness of the proposed method were illustrated by the optimization design of the scale-based universal motor families and the comparison against the designs obtained from the one-stage-Ps method.