Predicting the Lifetime of Flaky Tests on Chrome

Flaky tests not only disrupt the testing process but can also introduce significant delays in software release cycles, leading to increased costs and potential missed market opportunities. The ability to predict the lifetime class to which the test belongs to can empower development teams to allocate resources efficiently, prioritize bug fixes, and streamline the testing pipeline. Despite the critical role that this estimation plays in software development, there is a noticeable gap in the existing research literature. Bridging this gap is essential for enhancing testing efficiency and optimizing resource allocation and project planning in software development.In our investigation of the historical patterns of flaky tests in Chrome, we identified that 40% of flaky tests remain unresolved, while 38% are typically addressed within the initial 15 days of introduction. Subsequently, we developed a predictive model focused on identifying tests with quicker resolutions. Our model demonstrated a precision of 73% and a Matthews Correlation Coefficient (MCC) approaching 0.39 in forecasting the lifespan class of flaky tests.