基于偏好信息学习引导的混合性能指标智能优化决策模型与方法研究

项目名称： 基于偏好信息学习引导的混合性能指标智能优化决策模型与方法研究

项目编号： No.71201145

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： 张俊岭

作者单位： 浙江师范大学

项目金额： 19万元

中文摘要： 混合性能指标决策问题是一类广泛存在于管理优化决策领域中的复杂决策问题,具有决策目标难以完全数量化/结构化表示、决策者的偏好适应值具有噪声、问题决策空间大且可行方案数目多、群体决策等主要特征,需采用交互式决策机制,使得传统优化决策方法难以直接应用求解。本项目从决策模型与方法、问题处理流程机制与支持系统等方面深入研究基于偏好信息学习引导的混合性能指标智能优化决策模型、方法与关键支撑技术,具有重要理论意义,主要研究内容包括:①具有隐式性能指标适应值机器预测模型的交互式智能多目标优化决策模型②降低适应值"噪声"影响的单/多机器预测模型集成方法③有效提高决策绩效的自适应复合进化算子以及突破小种群限制的交互式智能多目标算法等关键支撑技术④基于诱导集结偏好信息学习引导的混合性能指标智能群体决策模型、方法及决策支持系统;所开发智能决策方法能够满足企业管理实践中提升核心竞争力的迫切需求,亦具有重要实际价值。

中文关键词： 混合性能指标优化问题；多属性决策；不确定性决策；智能优化；交互式决策

英文摘要： The hybrid-indices decision making problems widely residing in the enterprise operational optimization activities are complicately characterized by the main aspects as inclusion of unstructured tacit objectives,noise prefrence fitness of solutions, large search domain of solution space and the support of group decision making, which disables traditional optimizaion decision making methods and needs apporiate and effective interactive decision making methods. In this project, the intelligent interacitive optimization decision making approaches are studies through the following research outlines: ①intelligent interactive multiobjective optimization models with machine forecasting model for tacit fitness; ②integrated multiple-machine models for forecasting of tacit fitness to reduce the influence of noisy prefrence fitness of solutions; ③ critical techniques for effectively improving the efficiency of interactive decision making methods by research of adaptive hybrid evolutionary operators and intelligent interactive multiobjective optimization algorithms without the restriction of small population in interactive decision making environment; ④intelligent interactive group decision making models and methods guided by induced hybridly aggregated preference information and corresponding group decision support system.

英文关键词： Hybrid-indices optimization problems；multi-attribute decision making；uncertain decision making；intelligent optimization；interactive decision making