多目标非线性规划的可行方向交互算法

本文提出一个求解多目标非线性规划问题的交互规划算法．在每一轮迭代中，此法仅要求决策者提供目标间权衡比的局部信息．算法中的可行方向是基于求解非线性规划问题的Topkis－Veinott法构千的．我们证明，在一定条件下，此算法收敛于问题的有效解．

A Feasible Direction Interactive Algorithm for Mnltiobjective Nonlinear Programming Problems

An interactive algorithm is developed for solving multiobjective nonlinear programming problems. The algorithm requires the decision-maker to provide partial information, the local tradeoff ratios between objectives, in accordance with his utility function at each iteration. The feasible direction in this algorithm is constructed on the basis of the Topkis-Veinott method for solving nonlinear programming problems. Under certain conditions, the algorithm converges to an efficient solution of the discussed problem.