Fuzzy multi attribute decision making for the selection of UAV technology

Abstract

The rapid expansion of Unmanned Aerial Vehicle (UAV) technologies in both military and civilian domains has transformed these systems into essential components of modern operations. As UAV capabilities continue to evolve through artificial intelligence, advanced sensing, and autonomous control, they are now used in a wide range of missions such as surveillance, reconnaissance, logistics support, and precision strike. However, the diversity of available platforms and the mission-specific nature of performance requirements make UAV selection a complex multi-criteria decision problem. Each mission demands different operational priorities, and performance data often include uncertainty or expert subjectivity, creating a need for a structured and flexible evaluation framework. To address this challenge, the thesis introduces a Fuzzy Multi-Criteria Decision-Making (FMCDM) model that integrates Fuzzy Analytic Hierarchy Process (Fuzzy AHP) with Fuzzy TOPSIS. Unlike generic assessment approaches, the model evaluates UAV alternatives separately for three distinct military mission types, each with its own operational priorities. Expert evaluations were gathered using linguistic scales and pairwise comparisons, providing a fuzzy representation capable of capturing uncertainty and qualitative judgments. Five key performance criteria—payload capacity, operational range, endurance, maximum altitude, and maximum speed—were identified through literature review and expert consultation. Technical specifications of six anonymized UAV alternatives (UAV A–F) were converted into fuzzy linguistic forms to combine qualitative and quantitative information seamlessly. Fuzzy AHP was then used to calculate mission-specific criteria weights using Chang's extent analysis method, resulting in three different weight sets aligned with surveillance, combat, and logistics mission profiles. After determining the fuzzy weights, Fuzzy TOPSIS was applied to rank the UAV alternatives by measuring their closeness to the fuzzy ideal solution. Defuzzified results showed that the top-ranked UAV varied significantly across mission types, demonstrating that no single platform performs best in all operational scenarios. Some UAVs stood out in endurance-intensive reconnaissance missions, while others performed better in fast-response or payload-heavy mission settings. A comprehensive sensitivity analysis confirmed the robustness of the framework. While minor changes in criteria weights did not alter mission-specific rankings drastically, significant differences appeared across mission categories—supporting the need for dynamic, mission-driven evaluation rather than a uniform procurement strategy. Overall, the study offers a transparent, scalable, and practically applicable FMCDM model for UAV selection in defense planning. By integrating Fuzzy AHP and Fuzzy TOPSIS, the framework effectively handles both uncertainty and multi-dimensional performance data. Its adaptable structure allows for the inclusion of additional criteria such as autonomy level, stealth, maintenance cost, or geopolitical procurement risks. The thesis thus provides defense decision-makers with a scientifically grounded, mission-aligned tool for evaluating UAV alternatives and supports future developments involving group decision-making and real-time operational data.