Markovian Modeling of the Pix System Reliability Based on Real Incident Data

• Vitor Braço UFRJ
• Daniel Sadoc Menasché UFRJ
• Alberto Avritzer eSulabSolutions

DOI: https://doi.org/10.5753/wperformance.2025.8783

Abstract

Pix is a real-time payment system developed by the Central Bank of Brazil, requiring strict availability and performance guarantees. This paper proposes a Continuous-Time Markov Chain (CTMC) model to assess the reliability of Pix based on real incident reports. A dataset comprising operational failures was pre-processed and classified into five distinct system states, including normal operation and various failure modes. Transition rates were estimated empirically, and average recovery times were compared with official Service Level Agreements (SLAs) published by the Central Bank. Results indicate that observed recovery durations may exceed regulatory limits, highlighting the need for improved automation mechanisms.

References

Avritzer, A., Grottke, M., and Menasché, D. S. (2020). Using software aging monitoring and rejuvenation for the assessment of high-availability systems. In Handbook of Software Aging and Rejuvenation: Fundamentals, Methods, Applications, and Future Directions, pages 197–228. World Scientific.

Banco Central do Brasil (2022). Manual Operacional do DICT. Technical report, Banco Central do Brasil. [link].

Banco Central do Brasil (2023). Manual de Tempos do Pix - Versão 5.0. Technical report, Banco Central do Brasil. [link].

Banco Central do Brasil (2024). Relatório de gestão do pix – ano 2023. [link].

G1 (2023). Usuários reclamam de instabilidade no sistema do PIX nesta sexta-feira. [link].

Jacques-Silva, G., Avritzer, A., Menasche, D. S., Koziolek, A., Happe, L., and Suresh, S. (2015). Survivability modeling to assess deployment alternatives accounting for rejuvenation. In ISSREW, pages 114–119. IEEE.

Koutras, V. P. and Platis, A. N. (2020). On the performance of software rejuvenation models with multiple degradation levels. Software Quality Journal, 28.

Musa, J. D. (1975). A theory of software reliability and its application. IEEE Transactions on Software Engineering, 1(03):312–327.

Somesh, J. and Wing, J. M. (2001). Survivability analysis of networked systems. In Proceedings of the 23rd International Conference on Software Engineering. ICSE 2001, pages 307–317. IEEE.

Trivedi, K. S. (2001). Probability and statistics with reliability, queuing, and computer science applications. John Wiley & Sons.