INTELLIGENT MODEL OF INVENTORY MANAGEMENT BASED ON GENETIC ALGORITHMS AND FUZZY LOGIC FOR IMPROVING THE ECONOMIC SECURITY OF AN ENTERPRISE

Abstract

The article presents an intelligent inventory-management model that combines a genetic algorithm with fuzzy logic to enhance the economic security of an industrial enterprise. The genetic algorithm explores a large discrete solution space in search of near-global optima, whereas the fuzzy-logic module converts subjective assessments of supply-disruption risk into formalized penalties, thereby explicitly accounting for data uncertainty and incompleteness. The model incorporates hard constraints on resources, budget and production capacity, along with tunable soft penalties for product shortages, enabling a balance between cost minimization, contractual fulfilment and supplychain stability factors that directly influence the enterprise’s economic security. Experimental tests on a demonstration data set (three resource types and two product types) show that, after 60 generations of evolution, the system yields a profit of more than 37 thousand UAH while respecting all constraints and automatically optimizing purchasing and production volumes according to the risk profile. A sensitivity analysis confirms that modest adjustments to penalty weights allow rapid adaptation to strategic scenarios ranging from aggressive growth to conservative austerity. The proposed approach is readily scalable: adding new resources or products requires only updating the input parameters and fuzzy-inference rules, without restructuring the algorithm. The model can serve as the core of a digital twin or decision-support platform, providing managers with transparent analytical tools for scenario forecasting, operational-risk mitigation and reinforcement of economic resilience in turbulent markets and highly competitive environments.