Selection of an unmanned aerial vehicle for the Brazilian Navy: novel approach for decision support through unsupervised machine learning
Year: 2025
Type: article
Abstract: Purpose This paper describes the innovative MADRID Multicriteria Decision Analysis (MCDA) approach. This approach is used to assess Unmanned Aerial Vehicles (UAVs) and identify the best alternative for the Brazilian Navy to acquire for combat and support missions. Design/methodology/approach Based on the database available in de Araújo Costa et al. (2021), the UAVs were filtered by Endurance, Maximum speed, Maximum flight altitude, Maximum Take-Off Weight (MTOW), Payload and Cost. Findings The MADRID follows a structured approach to ranking alternatives. It selected the MQ-9 Reaper as the best UAV for the Brazilian Navy. Sensitivity analysis demonstrates the robustness and stability of the method under different weight configurations, ensuring the reliability of the results. Other high-performing alternatives are also highlighted, facilitating strategic decision-making in defense acquisitions. Future studies are suggested to include criteria related to the life cycle of UAVs. Research limitations/implications One key limitation of this study lies in the relatively small number of alternatives (14 UAVs), which may reduce the practical necessity of the clustering step in the MADRID methodology. While the method was designed for scalability and Big Data contexts, its full potential is best realized in larger decision spaces. Additionally, the results are based on a predefined set of operational criteria and expert evaluations, which, although robust, may vary under different strategic scenarios or stakeholder preferences. Originality/value MADRID integrates clustering techniques with MCDA methods, offering a solution for dimensionality reduction in Big Data problems and enhances decision-making by focusing on the most relevant alternatives. This methodology provides a robust and adaptable tool for strategic decisions, particularly in contexts with multiple evaluation criteria and large volumes of data. This approach is adaptable for acquiring a variety of new defense assets.
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: diamond
OpenAlex ID: W4411377441