不确定多目标满意决策问题及应用研究

项目名称： 不确定多目标满意决策问题及应用研究

项目编号： No.71201040

项目类型： 青年科学基金项目

立项/批准年度： 2013

项目学科： 管理科学与工程

项目作者： 纪颖

作者单位： 哈尔滨工业大学

项目金额： 19万元

中文摘要： 本项目将具有实际应用价值、也是决策领域的研究热点之一的不确定多目标决策作为研究的主要内容，研究基于满意度标准下的不确定多目标决策问题. 本项目试图从以下几个具体点出发开展研究.（1）提出一簇新的多目标满意度测量，新方法将能够同时兼顾现实决策的偏好性、决策的风险性以及分布的模糊性，同时这种方法将能够弥补现存的基于概率和期望值方法的不足.（2）将多目标满意度测量方法分别与鲁棒优化和分布鲁棒优化方法相结合,提出相应的基于多目标满意度测量的鲁棒优化和分布鲁棒优化问题，分析这两类最优化问题的最优条件，并构造求解这类问题的计算方法.（3）对不确定多目标投资组合管理及导弹作战过程中的不确定优化分配这两个实际问题，探讨其自身结构及模型特征，应用新方法求解并编制实用有效的计算机软件，并将新方法与基于概率的和期望值的方法进行比较.

中文关键词： 多目标决策；利率优化；供应链；近似点；算法

英文摘要： In this project, we will study the multiobjective satisficing decisionmaking problems under uncertain which has been widely used in practical and is one of the hot topics in decisionmaking field. In this project, we will consider this problem and its applications to the portfolio management problems and the Missile-Allocation problems under uncertainty, the contributions of this project are threefolds: (1) Firstly, a class of multiobjective sastisficing measure will be proposed which can not only balance the real decision preferences, the decision risk and distributional ambiguity, but overcome the shortage of the methods based on probability and mathematical expectation; (2) Secondly, the robust optimization model and the distributional robust optimization model under uncertainty will be given based on the multiobjective satisficing measures, respectively. The optimality conditions and algorithms will be proposed to analyze the resulted two optimization problems, respectively; (3) Finally, the new method will be applied to the portfolio management problem and the Missile-Allocation problem, respectively. The effectively numerical software will be presented. The comparisions to the methods based on probability and expectation will also be conducted.

英文关键词： multi-objective decision；portfolio optimization；supply chain；Proximal point；algorithm