https://ir.lib.vntu.edu.ua//handle/123456789/427 2026-04-05T17:26:30Z 2026-04-05T17:26:30Z Коваленко, В. П. Ковалюк, О. О. Kovalyuk, O. https://ir.lib.vntu.edu.ua//handle/123456789/50330 2025-12-26T11:15:53Z 2025-01-01T00:00:00Z

Аналіз використання штучного інтелекту для автоматизації процесів управління інфраструктурою підприємства Коваленко, В. П.; Ковалюк, О. О.; Kovalyuk, O. This paper research the application of artificial intelligence (AI) for automating enterprise infrastructure management processes. The increasing complexity of IT systems and the volume of operational data have rendered traditional monitoring and control approaches inefficient. In response, AI-driven solutions — particularly those involving machine learning (ML), deep learning (DL), and natural language processing (NLP) — have emerged as effective tools for enhancing reliability, adaptability, and operational efficiency. The study focuses on three core areas of AI integration: resource monitoring and optimization, automation of technical support, and intelligent document processing. In resource management, ML/DL algorithms outperform rule-based systems in anomaly detection, load forecasting, and fault prevention. The use of deep reinforcement learning enables adaptive resource allocation and improved fault tolerance in dynamic environments. In technical support, NLPbased chatbots can classify tickets, generate automated responses, and self-improve over time, significantly reducing mean time to resolution and enhancing user satisfaction. Intelligent document management systems, powered by neural networks and hybrid semantic-rule-based approaches, support automated classification, data extraction, and compliance checking in real time. To guide AI adoption, a structured methodology is proposed, comprising stakeholder identification, resource and data flow analysis, model selection, and staged integration. The paper introduces the concept of AI as a full-fledged participant in business processes rather than a passive tool. Within the Augmented Intelligence paradigm, AI agents and human experts collaborate—combining algorithmic precision with ethical reasoning and creativity. The practical implementation could demonstrate that AI integration leads to measurable improvements in efficiency, responsiveness, and decision quality across infrastructure-related workflows. Furthermore, it fosters a shift toward cognitive enterprise models, in which AI agents continuously learn and adapt to evolving business requirements. The findings offer a foundation for future research on human-AI collaboration, organizational transformation, and intelligent process design.; У статті проаналізовано можливості використання штучного інтелекту (ШІ) для автоматизації процесів управління інфраструктурою підприємства. Розглянуто ключові напрями впровадження ШІ: моніторинг ІТ-ресурсів, автоматизація технічної підтримки та інтелектуальне управління документообігом. Запропоновано алгоритм аналізу бізнес-процесів та впровадженні систем на основі ШІ. Особливу увагу приділено розгляду ШІ-агентів як учасників бізнеспроцесів.

2025-01-01T00:00:00Z Kovtun, V. Yasniy, O. Kovaliuk, O. Ковтун, В. В. Ковалюк, О. О. https://ir.lib.vntu.edu.ua//handle/123456789/50329 2025-12-26T11:19:39Z 2025-01-01T00:00:00Z

Markov model of controlled load distribution in the edge-IoT subsystem Kovtun, V.; Yasniy, O.; Kovaliuk, O.; Ковтун, В. В.; Ковалюк, О. О. The article presents a hybrid Markov model of adaptive load distribution within edge-IoT subsystems operating under conditions of stochastic traffic and heterogeneous environments. The proposed approach formalises the set of admissible and critical system states and introduces a routing policy with a probabilistic guarantee of stabilisation within QoS-defined configurations. For the first time, an integral efficiency criterion is proposed, combining the probability of remaining in desirable states with the minimum probability of stabilisation under perturbations. The model enables the development of adaptive routing strategies that minimise servicing costs and enhance resilience against node degradation and traffic fluctuations. Experimental results demonstrate the superiority of the proposed approach over classical strategies (round-robin and random) in terms of average delay, resource utilisation, and the loss function. The model has practical relevance for 6G edge architectures, autonomous systems, eHealth, and critical IoT infrastructures.

2025-01-01T00:00:00Z Білецький, Б. С. Biletskyi, B.S. https://ir.lib.vntu.edu.ua//handle/123456789/49977 2026-03-30T13:22:01Z 2022-01-01T00:00:00Z

Розроблення веб-системи обліку надання допомоги вимушено внутрішньо переміщеним особам в Україні Білецький, Б. С.; Biletskyi, B.S.

2022-01-01T00:00:00Z Стрембицький, П. Юхимчук, М. С. Лесько, В. О. Перепелиця, С. https://ir.lib.vntu.edu.ua//handle/123456789/44485 2025-07-18T13:46:37Z 2025-01-01T00:00:00Z

Централізований моніторинг інфраструктури з використанням системи Thanos: перспективи та виклики Стрембицький, П.; Юхимчук, М. С.; Лесько, В. О.; Перепелиця, С. The article presents a comprehensive study of the use of the Thanos system for centralized monitoring of IT infrastructure in the context of the growing complexity of modern information systems. The research introduces a systematic approach to building scalable monitoring solutions with focus on high availability and efficient resource utilization. A methodology for implementing and optimizing the monitoring system has been developed, encompassing all stages from data collection and processing to analysis and visualization. The study proposes and validates a mathematical model for evaluating system efficiency based on three key parameters: data compression coefficient (C), query processing performance (P), and system reliability (R), achieving an overall efficiency score of 0.9227. The experimental validation was conducted across three data centers (Kyiv, Lviv, Kharkiv) with 1,248 monitoring servers processing 127,492 metrics per second. The implementation demonstrates exceptional scalability with coefficients S(3) = 0.95 and S(5) = 0.94, while maintaining response times below 47ms under loads exceeding 900,000 queries per second. The system achieved significant improvements in key operational metrics: reducing incident response time from 15.7 to 9.4 minutes, increasing stored metrics volume from 2.3 to 23.7 petabytes while decreasing storage costs from $0.08 to $0.056 per gigabyte. The research analyzes practical applications across various sectors including banking, telecommunications, online retail, cloud computing, and healthcare, detailing specific monitoring requirements and implementation strategies for each domain. The study introduces innovative features including adaptive load balancing (LoadFactor = 0.6375) and intelligent caching (CacheEfficiency = 0.046). The paper concludes by outlining future development directions, focusing on machine learning integration for predictive analytics and enhanced automated incident response mechanisms; Стаття досліджує використання системи Thanos для централізованого моніторингу ІТ-інфраструктури. Проаналізовані поточні напрямки використання систем моніторингу та їх обмеження. Показано важливість централізованого моніторингу та можливості для подальшої інтеграції та розвитку системи Thanos.

2025-01-01T00:00:00Z