Product representation via networks methodology for exposing project risks

Shlomi Efrati, Yoram Reich*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


One of the significant factors in the time to market of a technology-based product development project is effective risk management. Both the system engineer and the project manager must work together to map and manage the risks in the project throughout its lifetime. Risks in the project arise from various reasons, which are not necessarily quantifiable, but all of which must be managed by the project team. We propose a methodology for calculating the risk level origin in each system element or component, taking into account its role within the system containing these elements and its availability in the project timeline. This risk level can be used as an additional decision support tool for the project stakeholders. For this purpose, we present a four-step process for (1) graph network mapping of products, (2) applying network algorithms, (3) weighting with information from the project management discipline, and (4) calculating risk index for identifying risks. The resulting level of risk index will enable the project team to map and manage efficiently and effectively the risks arising from the system components throughout the life of the development project. To demonstrate the methodology, we analyzed two products from different fields and at different levels of abstraction. We derived from each case the risk index for the use of the project personnel.

Original languageEnglish
Pages (from-to)421-442
Number of pages22
JournalResearch in Engineering Design - Theory, Applications, and Concurrent Engineering
Issue number4
StatePublished - Oct 2023


  • Betweenness centrality
  • Degree centrality
  • Network graph
  • Product design
  • Project manager
  • Risk management
  • System engineer


Dive into the research topics of 'Product representation via networks methodology for exposing project risks'. Together they form a unique fingerprint.

Cite this