Detecção de anomalias em processos licitatórios na Bahia : uma abordagem baseada em mineração de dados

Abstract

This study investigates the application of data mining techniques in detecting anomalies and irregularities in public procurement processes in the state of Bahia. The primary objective is to identify atypical patterns and indications of potential fraud, contributing to enhanced transparency and efficiency in public resource management. To achieve this, the research contextualizes legal aspects, with an emphasis on Law 14.133/2021, and analyzes the operational processes of public procurement, seeking to identify unusual behaviors that may indicate fraud. The methodology includes the use of advanced data analysis techniques, such as the Grubbs’ Test and the Local Outlier Factor (LOF), as well as other statistical methods, allowing the identification of suspicious patterns. Additionally, related studies and practical examples are explored to demonstrate the effectiveness of the adopted approaches. The analysis identified five procurement processes that stood out as anomalies, suggesting the need for further investigation into these specific cases. The expected results aim to contribute to the development of effective monitoring tools and the improvement of public management, enabling significant progress in preventing irregularities and optimizing procurement processes.