Reference Process for Integrating Data Science Workflows and Governance in Big Data Systems

This article presents a structured workflow for the development of data-based analysis in the public sector, from the identification of the institutional challenges to evidence-based decision-making. The proposed process organizes the steps around three main profiles - public manager, administrator and data scientist - and integrates data governance practices, LGPD compliance and reproducible analytical processes. The technical method for adding new analytics to the institutional platform is also detailed, involving the preparation of scripts, the use of machine learning models and the publication of analytical products with automated visualizations. The proposal contributes to standardization, transparency and efficiency in the adoption of analytical intelligence by public bodies, promoting evidence-based decisions.