PROBABILITY MODELS OF PRODUCTION INVENTORIES AT THE ENTERPRISE
Keywords:Inventories, Normal distribution, Central limit theorem
When managing production or commodity stocks, two main questions arise: when to replenish the stock and what should be its optimal size.
The purpose of this study is to build a probabilistic model, which can be proposed as a new inventory model, with the help of which the relationships between the period factors between the purchase of parts and their shelf life, which affect inventory management, are established.
Research methods are based on the approach using continuous distribution laws. The size of the reserve of parts is calculated depending on the established risk factor. Using the statistical method, point estimates were found for the studied parameters: average and root mean square deviation. A histogram of relative frequencies between the dates of two consecutive purchases is constructed. Critical areas for the studied parameters are illustrated. The value of the difference in days between the purchase of parts and the amount of the purchase of parts, which correspond to the normal laws of the distribution of random variables with the appropriate parameters, as well as the critical values of the need for parts in the production process, were calculated. The size of the parts reserve was found, which corresponds to the normal distribution law, depending on the established risk factor. For different values of this coefficient, the value of the difference in days between purchases of parts, the amount of purchases and the reserve of parts, which correspond to the distributions of random values, as well as the critical value of the need for parts in the production process to avoid production downtime are given. Using the central limit theorem, it is shown that the purchase volume of parts and the volume of used parts are normally distributed.
Taking into account the degree of uncertainty associated with the structure of demand and the time of use of stocks at the enterprise, the authors chose probabilistic models that make it possible to flexibly change simulated demand and take this into account in forecasting.
The research concluded that the probabilistic approach is the basis for forecasting inventory management at the enterprise, which takes into account the risks associated with determining the optimal need for raw materials at the enterprise.