Modification of Algorithms for Forecasting the Performance of Corporate Information Systems Using Artificial Intelligence

Abstract

The aim of the study is to improve the algorithm for forecasting the performance of corporate information systems in order to take into account the non-linear nature of workload dynamics and ensure a timely identification of hidden patterns in digital infrastructure. The relevance of the study is determined by the growing complexity of corporate information systems, increasing data volumes and the need to prevent critical states before they affect business processes. The methodology is based on system analysis, a hierarchical representation of monitoring processes, intelligent data processing methods and decision-making theory. The article proposes a comprehensive forecasting architecture that combines telemetry collection, feature space formation, adaptive machine learning, anomaly detection and proactive resource management. The scientific and practical significance of the findings lies in creating a theoretical basis for autonomous adaptive control systems capable of maintaining stable operation under changing load conditions.