Forecasting the Success of Projects Based on the Analysis of Structural Characteristics of Communication Networks

The article presents a methodology for predicting the success of project initiatives based on an analysis of the structural characteristics of project team communication networks. The study is based on data from task tracking systems (Jira, Trello) that reflect formal interactions between project participants. The research methodology includes a set of analytical tools: correlation analysis using Spearman’s and Pearson’s coefficients, regression modeling based on the Random Forest algorithm, and anomaly detection methods using Isolation Forest. The study revealed statistically significant correlations between key network metrics (betweenness centrality, network density, graph diameter) and project performance indicators (adherence to deadlines, budget, quality of results). A statistically significant negative correlation was found between excessive centralization and adherence to deadlines (ρ = –0.72), as well as a positive correlation between network density and quality of results (r = 0.68). The developed model based on Random Forest demonstrates the accuracy of forecasting the success of projects at the level of 84%. It was found that excessive centralization of communications reduces the probability of successful project implementation, while the optimal density of the communication network contributes to the achievement of project KPIs. The practical significance of the study lies in the possibility of early detection of project failure risks based on objective metrics of communication activity. The developed methodology, tested on IT company data, allows not only to predict risks, but also to form recommendations for optimizing team interactions. The results of the study are of interest to project managers, HR analysts, and data-driven management specialists.

1. Kohn M. Agile: Project assessment and planning. Transl. from Eng. Moscow: Alpina Publisher; 2022. 418 p. URL: https://stroki.mts.ru/book/agile-otsenka-i-planirovaniye-proyektov-1759?ysclid=mdhbufcngh64422956 (In Russ.).

2. Durland M., Fredericks K. An Introduction to Social Network Analysis. New Directions for Evaluation. 2006;2005(107):5–13. DOI: 10.1002/ev.157

3. Otte E., Rousseau R. Social network analysis: A powerful strategy, also for the information sciences. Journal of Information Science. 2002;28(6):441–453. DOI: 10.1177/016555150202800601

4. Krackhardt D., Hanson J. R. Informal networks: The company behind the charts. Harvard Business Review. 2000;78(4):104–11. URL: https://www.researchgate.net/publication/40961009_Informal_Networks_the_Company_behind_the_Charts

5. Pryke S. D. Towards a social network theory of project governance. Construction Management and Economics. 2005;23(9):927–939. DOI: 10.1080/01446190500184196

6. Chinowsky P., Diekmann J. and Galotti V. Social network model of construction. Journal of Construction Engineering and Management. 2008;134(10). DOI: 10.1061/(ASCE)0733-9364(2008)134:10(804)

7. Titov S.A. Investigation of the relationship between the structure of intra-project communications and the effectiveness of innovative projects using the analysis of social networks. Cloud of Science. 2014;1(4):665–695. URL: https://www.elibrary.ru/thbiwr (In Russ.).

8. Plokhov D.V. et al. Methodology for assessing the impact of social communications on the effectiveness of an innovation project. Cloud of Science. 2016;3(3). URL: https://www.researchgate.net/publication/309788251_Metodika_ocenki_vliania_socialnyh_kommunikacij_na_rezultativnost_innovacionnogo_proekta (In Russ.).

9. Ilin I., Voronova O., Pavlov D. et al. System of project management at a medical hub as an instrument for implementation of open innovation. Systems. 2023;11(4):182. DOI: 10.3390/systems11040182

10. Pavlov D.A., Kiri A.V. Methodology of monitoring and control of innovative projects by methods of social network analysis. Polythematic online electronic scientific journal of the Kuban State Agrarian University. 2021;166:169–181. (In Russ.). DOI: 10.21515/1990-4665-166-015

11. Pavlov D. A., Kiri A.V. Development of a set of problem-oriented programs for solving problems of network distribution of production tasks. Proceedings of the Kuban State Agrarian University. 2020;86:26–30. URL: http://opac.skunb.ru/index.php?url=/notices/index/IdNotice:1396389/Source:default&ysclid=mdhdf669fe544643381# (In Russ.).

12. Rashka S., Mirjalili V. Python and machine learning: machine and deep learning using Python, scikitlearn and TensorFlow 2. 3rd ed. Transl. from Eng. St. Petersburg: Dialectics; 2020. 848 p. URL: https://www.williamspublishing.com/Books/978-5-907203-57-0.html (In Russ.).

13. Pavlov D.A., Kiri A.V. Forecasting the success of IT projects. Polythematic online electronic scientific journal of the Kuban State Agrarian University. 2024;203:505–514. (In Russ.). DOI: 10.21515/1990-4665-203-045