SEQUENCE OF DEVELOPING INVENTORY MANAGEMENT OPTIMIZATION AT LIGHT INDUSTRY ENTERPRISES

Abstract

The article investigates the problem of inventory management optimization at light industry enterprises as a complex multi-criteria managerial task involving the continuous alignment of economic interests and production constraints. It is established that the classical dilemma between inventory surplus, which freezes working capital and increases storage costs, and material shortage, which causes production downtime and financial losses, is particularly acute in light industry due to three compounding factors: demand seasonality with production volume fluctuations two to three times above average levels, a wide assortment of materials of various colors, textures and compositions, and high variability of consumer preferences driven by fashion trends, where model life cycles span only three to four months. An additional systemic factor complicating the implementation of modern management tools is the insufficient level of digital maturity of enterprises in the industry, defined as the degree of development and integration of digital technologies into business processes, which necessitates a differentiated approach to selecting methods and technologies depending on the scale and technological readiness of the enterprise. It is substantiated that traditional inventory management methods developed for industries with relatively stable demand and limited product assortments prove insufficiently effective without a systematic approach, where ABC analysis without prior diagnostics remains a formality and an ERP system without clearly defined goals becomes unjustified expenditure.

A five-stage inventory management optimization sequence with feedback mechanisms between stages is proposed, ensuring cyclical adaptation of the system to changes in market conditions and fashion trend dynamics. A matrix of differentiated strategies for nine material groups based on ABC-XYZ analysis is developed, defining control frequency, management method and safety stock level for each group, where daily control with precise EOQ calculations is recommended for high-value stable-demand materials of group AX, while simplified visual control and order-per-project approaches are sufficient for low-value unpredictable-demand materials of group CZ. A three-level digitalization system is substantiated: the basic level with automatic material write-off based on technological maps is sufficient for small enterprises with up to 500 material items; the medium level with ERP-CAD integration that automatically calculates material requirements based on collection designs is appropriate for enterprises with 500–3000 items; the advanced level incorporating AI trend forecasting with accuracy of 90–95% compared to 60–70% for traditional methods and IoT sensors providing automatic inventory tracking with 99.7% accuracy is justified for large enterprises with over 3000 items. A comprehensive multi-criteria inventory management model, combining an industry-oriented diagnostics block covering six diagnostic directions, multi-criteria material classification integrating economic, statistical and industry-specific criteria including production role, seasonality and life cycle type, a differentiated strategies system, a parametric calculation subsystem accounting for seasonality and model life cycles, technological-digital integration ensuring real-time data support, and an adaptive feedback mechanism. The cyclical nature of the model ensures its continuous adaptation to the dynamics of the fashion industry, making it applicable to light industry enterprises of varying scale and assortment complexity.