Defesa de Dissertação: Pedro Paulo Cortez
Data da publicação: 29 de julho de 2025 Categoria: Defesas de Dissertação, Notícias
Título: The Impact of Continuous Practices on Software Maintenance and Evolution: An Industrial Case Study
Data: 05/08/2025
Horário: 09h
Local: Google Meet (https://meet.google.com/iax-khck-qkn)
Resumo:
Context. Continuous practices (e.g., CI/CD, observability, automated testing) are widely promoted as ways to enhance software delivery and quality. However, evidence about their impact in legacy-constrained and regulated environments remains inconclusive.
Objective. This study investigates how continuous practices affect both delivery performance and team experience in a long-lived mission-critical system from industry. We aim to understand not only the technical outcomes but also the organizational changes that happen along the way.
Method. We conducted a mixed-methods case study combining quantitative analysis of development tasks with qualitative insights from structured interviews. The analysis included time-based metrics (e.g., active working time, completion time), descriptive statistics, and hypothesis testing. Outliers were removed using an interquartile range (IQR) strategy. The interviews explored perceptions of challenges, software quality, and workflow evolution.
Results. Statistical hypothesis testing did not indicate significant differences in any of the evaluated time metrics. However, a geometric decreasing trend was observed in both active working time and time to completion following the adoption of continuous practices, along with reduced variance. Interview responses further revealed that continuous practices improved the perceived software quality and predictability.
Conclusions. The study provides empirical evidence on the effects of adopting continuous practices in a real-world, large-scale, legacy-constrained software project. Beyond tools, success depends on how well these practices fit with existing workflows and constraints. We hope this case contributes practical lessons for teams facing similar modernization journeys.
Objective. This study investigates how continuous practices affect both delivery performance and team experience in a long-lived mission-critical system from industry. We aim to understand not only the technical outcomes but also the organizational changes that happen along the way.
Method. We conducted a mixed-methods case study combining quantitative analysis of development tasks with qualitative insights from structured interviews. The analysis included time-based metrics (e.g., active working time, completion time), descriptive statistics, and hypothesis testing. Outliers were removed using an interquartile range (IQR) strategy. The interviews explored perceptions of challenges, software quality, and workflow evolution.
Results. Statistical hypothesis testing did not indicate significant differences in any of the evaluated time metrics. However, a geometric decreasing trend was observed in both active working time and time to completion following the adoption of continuous practices, along with reduced variance. Interview responses further revealed that continuous practices improved the perceived software quality and predictability.
Conclusions. The study provides empirical evidence on the effects of adopting continuous practices in a real-world, large-scale, legacy-constrained software project. Beyond tools, success depends on how well these practices fit with existing workflows and constraints. We hope this case contributes practical lessons for teams facing similar modernization journeys.
Banca examinadora:
- Prof. Dr. Lincoln Souza Rocha (MDCC/UFC) – Orientador
- Prof. Dr. Camilo Camilo Almendra (UFC) – Coorientador
- Prof. Dr. Fernando Antônio Mota Trinta (MDCC/UFC)
- Prof. Dr. Guilherme Horta Travassos (UFRJ)
