摘要
针对项目组合中的风险分析问题,本文通过构建"项目组合-风险"双层网络模型,提出了基于随机游走算法的项目组合风险定量分析方法,以更准确地预测项目组合风险.首先,本文在研发项目间依赖关系分析的基础上,构建了项目组合网络,并采用设计结构矩阵(DSM)法建立了基于复杂网络中"邻接节点"的项目间连接强度模型.然后,采用多领域矩阵建立了由风险事件推导风险因素间依赖关系的模型.进一步,通过集成项目组合中项目间依赖关系、项目与风险因素间对应关系、以及风险因素间初始的依赖关系,构建了"项目组合-风险"双层网络模型,采用随机游走算法得到稳定状态下最终的"项目-风险因素"域映射矩阵(DMM)和风险因素DSM.最后,以某研发项目组合为例,采用随机游走算法进行项目组合风险的精准预测,并采用PageRank方法对风险因素进行排序,验证了本文提出模型和方法的有效性.
To solve the risk analysis problem in the project portfolio,the paper builds the two-layer"project portfolio-risk"network and proposes to use the restart random walk algorithm for the project portfolio risk analysis.The quantitative analysis method of portfolio risk proposed in this paper can more accurately assess the portfolio risk and rank the risk factors.Firstly,based on the analysis of the independency among projects,we build the network of the project portfolio,then we use the design structure matrix(DSM)and propose a model to measure the tie strength among projects based on the"adjacent nodes"in the portfolio network.Then,using multi-domain matrix(MDM),the inference model of dependency relationship among risk factors is built,which can be used to derive the relationship between risk factors from risk event.Next,the paper builds the two-layer"project portfolio-risk"network which integrates the interdependency relationship among projects,the initial interdependency relationship among risks factors and the corresponding relationship among the projects and risk factors.Further,the paper applies the restart random walk algorithm to build the portfolio risk analysis model.Finally,an industrial example is provided to illustrate the validity of proposed models.
作者
杨青
邹星琪
李兆超
YANG Qing;ZOU Xingqi;LI Zhaochao(Donlinks School of Economics and Management,University of Science and Technology Beijing,Beijing 100083,China)
出处
《系统工程理论与实践》
EI
CSSCI
CSCD
北大核心
2020年第7期1832-1843,共12页
Systems Engineering-Theory & Practice
基金
国家自然科学基金(71872011,71929101)
中央高校基本科研业务费(FRF-BR-18-001B)。
