Matías López San Martín, David Ramon Núñez Benitez, Julio Manuel Paciello Coronel, Juan Ignacio Pane Fernandez

Proceedings of the 15th International Multi-Conference on Complexity, Informatics and Cybernetics: IMCIC 2024, pp. 164-169 (2024); https://doi.org/10.54808/IMCIC2024.01.164

Matías López San Martín
Polytechnic Faculty, National University of Asuncion, San Lorenzo, Central, Paraguay

David Ramon Núñez Benitez
Polytechnic Faculty, National University of Asuncion, San Lorenzo, Central, Paraguay

Julio Manuel Paciello Coronel
Polytechnic Faculty, National University of Asuncion, San Lorenzo, Central, Paraguay

Juan Ignacio Pane Fernandez
Polytechnic Faculty, National University of Asuncion, San Lorenzo, Central, Paraguay

Cite this paper as:
López San Martín, M., Núñez Benitez, D. R., Paciello Coronel, J. M., Pane Fernandez, J. I. (2024). Quantifying the Risk of Complaints in Public Procurement Tenders in Paraguay Using Machine Learning. In N. Callaos, S. Hashimoto, N. Lace, B. Sánchez, M. Savoie (Eds.), Proceedings of the 15th International Multi-Conference on Complexity, Informatics and Cybernetics: IMCIC 2024, pp. 164-169. International Institute of Informatics and Cybernetics. https://doi.org/10.54808/IMCIC2024.01.164

Public procurement processes are critical for effective governance, and they are susceptible to protests, causing delays and possibly added costs. This study aims to assess the risk of protests in paraguayan public procurement tenders using artificial intelligence. The research leveraged machine learning techniques, including a classifier, to analyze historical data from the Public Procurements Office of Paraguay (DNCP for its initials in Spanish) available in the open format of the Open Contracting Data Standard (OCDS). Pre-calculated red flags were incorporated into the model as an indicator of potential irregularities. A structured and unified dataset was generated, laying the foundation for future investigations. The model exhibited promising predictive capabilities identifying the procurement tenders at high risk of protests. This work represents a significant step towards proactive protest risk management in public procurement. The combination of complaint-derived data, machine learning, and the structured dataset enhances the potential for technology-driven transparency and efficiency in public procurement processes.