一类不可微广义凸多目标规划的最优性条件和对偶

摘要

The paper discusses the non-differentiable multiobjective programming problem with inequality constraints. For this class of problems, the concepts of the generalized d- type-I univexity function are introduced, and Karush-Kuhn-Tucker-type sufficient optimality conditions are proved. Moreover, a mixed duality model is constructed and duality theorems are established.%本文研究带不等式约束的不可微多目标规划问题,引入了广义d-I型一致不变凸函数的概念,证明了Pareto有效解和Pareto弱有效解的Karush-Kuhn-Tucker型充分条件.构造出了混合对偶模型,并证明了相应的对偶定理.