信息不确定下的主动打断项目组合选择问题鲁棒优化

发布时间：2018-05-22 13:20

本文选题：项目组合选择 + 鲁棒优化　；参考：《系统工程理论与实践》2017年11期

【摘要】：在项目组合选择问题中,历史数据的缺乏以及预测和估计过程中出现的不可避免的误差,会导致模型中的参数无法被准确地估计,进而给决策带来巨大的风险.因此,构建合适的鲁棒优化模型,为企业提供能有效应对参数不确定性的鲁棒解,对企业的风险防范具有极其重要的现实意义.本文首先对确定参数下的主动打断项目组合选择问题数学模型的特点进行了分析.进一步地,介绍了鲁棒优化问题中不确定情境集的概念,并给出了允许管理者根据其偏好确定不确定情境集大小的方法,构建了全新的基于情境的鲁棒优化模型,进而计算出在所规定的不确定情境集内的最坏情境下能保持可行性与最优性的鲁棒解,实现了鲁棒性与最优性间的权衡,最后,通过GAMS/BARON进行了算例分析,验证了模型的合理性与有效性.从理论上,本文首次将鲁棒优化理论扩展到了主动打断项目组合选择问题中,针对现有的项目组合选择问题鲁棒优化理论仅能应对有限个可行解的不足之处,提出了一类新的鲁棒优化方法,使其能够应对具有无穷多可行解的主动打断项目组合问题.从实践上,随着我国高新产业的发展,具有超前性与特殊性的研究与发展(RD)、信息科技与信息系统(IT/IS)等新兴项目的投资日益受到重视.相较于传统项目,这类项目的高度不确定性使得探究项目组合选择问题的鲁棒优化理论日益迫切.故而本文的研究具有明显的理论价值和现实意义.
[Abstract]:In the problem of portfolio selection, the lack of historical data and the inevitable errors in the process of prediction and estimation will lead to the inability of the parameters in the model to be accurately estimated, and thus bring great risks to the decision. Therefore, a suitable robust optimization model is constructed to provide enterprises with an effective response to the uncertainty of parameters. It is of great practical significance to the risk prevention of enterprises. Firstly, this paper analyzes the characteristics of the mathematical model of the active interruption of the project portfolio selection problem under the determination of parameters. Further, the concept of the uncertain situation set in the robust optimization problem is introduced, and the uncertainty is given to the managers to determine the uncertainty according to their preference. A new scenario based robust optimization model is constructed, and then the robust solution that can maintain the feasibility and optimality in the worst situation set is calculated. The tradeoff between robustness and optimality is realized. Finally, an example analysis is carried out by GAMS/BARON to verify the combination of the model. In theory, the robust optimization theory is extended to the problem of active interruption of portfolio selection for the first time. For the existing combination selection problem, the robust optimization theory can only deal with the shortcomings of the finite feasible solutions. A new kind of robust optimization method is proposed to deal with the infinitely many feasible methods. In practice, with the development of high and new industries in China, the research and development of advanced and special features (RD), information technology and information system (IT/IS) and other emerging projects have been paid more and more attention. Compared with traditional projects, the high uncertainty of these projects makes the inquiry project portfolio selection The theory of robust optimization is becoming more and more urgent. Therefore, the research in this paper has obvious theoretical and practical significance.
【作者单位】：华北电力大学经济与管理学院;北京航空航天大学经济管理学院;北京市科学技术情报研究所;
【基金】：国家自然科学基金(71471006,71772060) 北京市财政项目:情报快速生产线建设(Ⅰ期)(PXM2017_178214_000005)~~
【分类号】：F224;F276.44

【相似文献】