Creating Actionable and Insightful Knowledge Applying Graph-CentralityMetrics toMeasure Project Collaborative Performance

Abstract

Tools and techniques supported by math and statistics are often used by organizations to measure performance. These usually measure an employees’ traits and states performance. However, the third type of data usually neglected by organizations, known as relational data, can provide unique and actionable insights regarding the root causes of individual and collective performance. Relational data are best captured through the application of graph-based theory due to its ability to be easily understood and quantitatively measured, while mirroring how employees interact between them as they perform work-related tasks or activities. In this work, we propose a set of graph-based centrality metrics to measure relational data in projects by analyzing the five most voted relational dimensions ((1) communication, (2) internal and external collaboration, (3) know-how exchange and informal power, (4) team-set variability, and (5) teamwork performance), in a survey conducted to 700 international project stakeholders in eight business sectors. The aim of this research is to tackle two issues in projects: First, to understand in a quantitative way how the project’s relational data may correlate with project outputs and outcomes, and second, to create unique and actionable knowledge to help mitigate the increasing project failure rates. A case study illustrates the step-by-step application of the developed graph-based metrics as well as its benefits and limitations.