Multiobjective engineering design optimization problems: a sensitivity analysis approach

Pesquisa Operacional

Oscar Brito Augusto²  Fouad Bennis¹  Stephane Caro¹ 

[1] ,Universidade de São Paulo Escola Politécnica

关键词: multiobjective optimization;    Pareto-optimal solutions;    sensitivity analysis;   

DOI  :  10.1590/S0101-74382012005000028

来源: SciELO

【 摘 要 】

This paper proposes two new approaches for the sensitivity analysis of multiobjective design optimization problems whose performance functions are highly susceptible to small variations in the design variables and/or design environment parameters. In both methods, the less sensitive design alternatives are preferred over others during the multiobjective optimization process. While taking the first approach, the designer chooses the design variable and/or parameter that causes uncertainties. The designer then associates a robustness index with each design alternative and adds each index as an objective function in the optimization problem. For the second approach, the designer must know, a priori, the interval of variation in the design variables or in the design environment parameters, because the designer will be accepting the interval of variation in the objective functions. The second method does not require any law of probability distribution of uncontrollable variations. Finally, the authors give two illustrative examples to highlight the contributions of the paper.

【 授权许可】

CC BY   
 All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License