A integração entre mineração de dados e gestão do conhecimento como suporte à tomada de decisão na gestão de pessoal da Marinha do Brasil

Abstract

The contemporary scenario is characterized by a significant increase in data production and availability, which poses to organizations the challenge of converting them into relevant information and applicable knowledge for decision-making processes. Within the Brazilian Navy (MB), career management of military personnel, conducted by the Directorate of Naval Personnel (DPM), involves high complexity due to the integrated analysis of multiple variables related to military trajectories. Among the aspects that require attention, attrition stands out, as it influences career flow, personnel sizing, and the preservation of organizational knowledge. In this institutional context, Knowledge Management (KM) encompasses structured processes to create, organize, share, and apply organizational knowledge, ensuring that strategic information is available at the right time to support decisions. Complementarily, Data Mining (DM) applies statistical and computational methods to identify patterns and trends in large datasets, providing analytical insights that can enhance the accuracy and consistency of choices. The integration of KM and DM, therefore, emerges as a promising alternative to transform available historical data into applicable knowledge, contributing to the quality and reliability of personnel management decisions. This study aimed to analyze the potential application of DM techniques in career planning within the MB, with emphasis on predictive classification models to identify profiles with higher propensity for attrition. The research, of bibliographic and descriptive nature, addressed DM tasks and techniques, the institutional context, and the comparison of classification methods. As a result, decision trees proved to be suitable for the problem, as they combine predictive accuracy with interpretability, enabling the identification of critical variables and supporting more effective retention policies. It is expected that the adoption of this model will strengthen decision-making, optimize personnel management, and contribute to the institutional efficiency of the MB.