查询结果:   易高明,蒋艳.基于目标相对重要性的模糊多目标进化算法[J].计算机应用与软件,2018,35(6):272 - 278.
中文标题
基于目标相对重要性的模糊多目标进化算法
发表栏目
算法
摘要点击数
1131
英文标题
FUZZY MULTI-OBJECTIVE EVOLUTIONARY ALGORITHM BASED ON THE RELATIVE IMPORTANCE OF TARGETS
作 者
易高明 蒋艳 Yi Gaoming Jiang Yan
作者单位
上海理工大学管理学院 上海 200082     
英文单位
Business School, University of Shanghai for Science and Technology, Shanghai 200082, China     
关键词
多目标优化 决策者 目标相对重要性 模糊偏好
Keywords
Multi-objective optimization Decision makers The relative importance of targets Fuzzy preferences
基金项目
作者资料
易高明,硕士生,主研领域:智能算法,多目标优化。蒋艳,副教授。 。
文章摘要
针对决策者实践需要与算法有效资源分配不均衡的多目标优化问题,提出一项更加符合人的认知规律的表达目标间重要性的九级语义范式,构建偏好区域以交互式的刻画决策者认知的模糊性与渐进性来引导进化种群向偏好域内逼近,进而给出一种模糊锦标赛选择算子——RPT-dominance算子。该交互式偏好嵌入算法在NSGA-II框架下进行操作。将该方法应用在五个标准测试函数上并与偏好先验嵌入法、偏好后验嵌入法作对比实验,结果表明该算法在处理模糊偏好的多目标问题上具有较显著的优势。
Abstract
For the multi-objective optimization problem in which decision makers need to implement an imbalanced allocation of effective resources, a nine-level semantic paradigm that is more in line with the cognitive laws of human beings was proposed. The construction preference region guided the evolutionary population to approach in the preference domain by interactively portraying the vagueness and gradualness of decision makers cognition, and then presented a fuzzy tournament selection operator——RPT-dominance operator. The interactive preference embedding algorithm operated under the NSGA-II framework. The method was applied to five standard test functions and compared with the preferred prior embedded method and the preferred posterior embedded method. The results show that the algorithm has significant advantages in dealing with multi-objective problems of fuzzy preferences.
下载PDF全文