计算机工程与应用 ›› 2022, Vol. 58 ›› Issue (2): 265-273.DOI: 10.3778/j.issn.1002-8331.2007-0398

• 工程与应用 • 上一篇    下一篇

动态供应链网络中企业合作关系的链路预测

卢志刚,陈倩   

  1. 上海海事大学 经济管理学院,上海 201306
  • 出版日期:2022-01-15 发布日期:2022-01-18

Link Prediction of Enterprise Cooperation Relationship in Dynamic Supply Chain Network

LU Zhigang, CHEN Qian   

  1. College of Economics and Management, Shanghai Maritime University, Shanghai 201306, China
  • Online:2022-01-15 Published:2022-01-18

摘要: 随着经济全球化的推进,供应链由传统的单链形式逐渐演变成供应链网络。这类复杂供应链网络由不同类型的企业构成,为了适应外部环境的变化,企业与企业之间以供需为目的形成的合作关系随着时间不断改变。致力于动态供应链网络中企业合作关系的预测,提出一种基于投影和时间事件的链路预测模型,将动态供应链网络划分为由若干个时间片组成的网络快照,通过对网络快照进行投影分析,形成潜在合作链路集合,基于时间事件计算集合里每个元素的链路预测得分,比较得分确定预测结果。实验分析结果表明,相对传统链路预测算法,所提算法在动态供应链网络中预测企业间未来合作关系具有更好的预测效果,且通过改变网络快照的大小和时间事件变化率可以进一步提高其预测精度。

关键词: 动态供应链网络, 合作关系, 链路预测

Abstract: With the advancement of economic globalization, the supply chain has gradually evolved from a traditional single-chain form to a supply chain network. This type of complex supply chain network is composed of different types of enterprises. In order to adapt to the changes in the external environment, the cooperation relationship formed with enterprises for the purpose of supply and demand continues to change over time. Committed to the prediction of enterprise cooperation relationship in the dynamic supply chain network, a link prediction model based on projections and temporal events is proposed. The dynamic supply chain network is divided into network snapshots consisting of several time slices. Through projection analysis of the network snapshots, a set of potential cooperation links is formed. The prediction score of each element in the set is calculated based on the temporal event. Comparing the prediction scores, the results are attained. Experimental analysis results show that, compared with traditional link prediction algorithms, the proposed algorithm has a better prediction effect in predicting the future cooperation relationship between enterprises in a dynamic supply chain network, and it is concluded that the prediction precision can be further improved by changing the size of the network snapshot and the change rate of temporal events.

Key words: dynamic supply chain network, cooperation relationship, link prediction