This study aims to identify possible indications of irregularities in public procurement processes, enabling an assessment of their feasibility within oversight and auditing procedures. For a better understanding of this study, a literature review was conducted in order to examine the facts and phenomena related to public procurement. Concepts of analytical intelligence were explored, including Business Intelligence, Data Science, and data mining techniques. Subsequently, the data selection process was initiated, and suitable datasets made available through the transparency portal of the Office of the Comptroller General of the State were identified. The dataframes related to the institution’s procurement processes were structured using data associated with the period from January 2016 to December 2019. Through descriptive analysis using Microsoft Power BI, it was possible to achieve a broader understanding of the variables involved and to identify indications of patterns within the data. The predictive analysis that followed enabled the generation of rules associating two or more companies with joint participation and winning outcomes, which may indicate potential irregularities. The elements presented demonstrate significant potential for the development of decision-support systems and strategies aimed at monitoring public procurement processes, as they generate information that allows greater accuracy in identifying possible irregularities, thereby improving the efficiency of public resource allocation and utilization.
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Copyright (c) 2021 Brazilian Journal of Criminalistics
https://orcid.org/0000-0002-6747-1822
Escola de Ciências Exatas e de Computação, Pontifícia Universidade Católica de Goiás, Goiânia (GO), Brasil