一类不可微广义分式规划的最优性条件

Optimality conditions for a class of nondifferentiable minimax fractional programming

提出了一类目标函数的分子和分母中都含有支撑函数的不可微广义分式规划问题,在Kuhn-Tucker约束品性下,给出了这类广义分式规划的Kuhn-Tucker型必要条件,并在不可微函数的广义(F,ρ)-凸性假设下,给出了它的最优性充分条件,所提的问题及所得结果相对现有文献具有一般性。

A class of nondifferentiable minimax fractional programming problem is put forward, in which the numerator and the denominator of the objeetive function involve support function. Under Kuhn-Tucker constraint qualification, the Kuhn-Tucker type necessary conditions are given. The sufficient optimality condition is also given with generalized （F,p）-convexity assumptions. The problems and the results in this paper are more general than the relevant ones appeared before.