A modern approach to calculating the abilities and capabilities of military units to perform tasks using artificial intelligence

• Datum: 2025
• Hauptverfasser: Бойченко, А. В., Додонов, В. О., Залужний, В. Ф., Ізварін, Є. І.
• Format: Artikel
• Sprache: Ukrainisch
• Veröffentlicht: Інститут проблем реєстрації інформації НАН України 2025
• Schlagworte: neural network; capability; readiness; artificial intelligence
• Online Zugang: http://drsp.ipri.kiev.ua/article/view/345668
• Назва журналу: Data Recording, Storage & Processing

The research examines the issue of assessing the capabilities of military units, which is highly relevant in modern conditions. Although formula-based evaluation methods are still used today, they require improvement, particularly in refining results and increasing calculation speed. The advantages of this approach include ease of use, the absence of a need for large volumes of input data, minimal computational resource requirements, as well as the transparency and clarity of the algorithms. However, this approach also has significant limitations, as it covers a narrow set of factors and cannot adapt to changes in the combat environment. A problem arises when the calculated value falls on the borderline between capabi-lity categories. In such cases, the risk of making erroneous managerial decisions regarding a unit's ability to execute a combat mission increases. A system using neural networks for capability assessment is proposed, which would significantly enhance the accuracy, objectivity, and efficiency of calculating the capabilities and readiness of military units. Since the principle of capability assessment has a hierarchical structure, an effective solution would be to create a tree-like or network-centric architecture utilizing multiple interconnected neural networks, each responsible for calculating a specific capability it was trained for. Such a system could compute multiple capabilities simultaneously, greatly improving the speed at which command receives objective information — something the formula-based approach cannot achieve. The results of these calculations could be integrated into military analytics systems, enhancing situational awareness for command and helping to avoid knowingly unfeasible missions. Tabl.: 1. Fig.: 5. Refs: 14 titles.