206 満足化トレードオフ法による多目的最適化問題の厳密解法と応用

抄録

Satisficing Trade-off Method(STM) is one of the effective methods to deal with multi-objective optimization problems(MOOP), solving the optimality conditions exactly. In MOOPs, Multi-Objective Genetic Algorithm(MOGA) has became very popular nowadays, however, it is a kind of heuristic algorithm and no guarantee of optimality on their solutions. In this point of view, STM has significant value in MOOP, especially in quantitative design consideration. In this paper, first we will discuss about STM's numerical characteristics and show our real implementation in general comercial software iSIGHT 10.0. Then, its application and use scenario will be discussed comparing with MOGA.