Proper management of inventories within any company is important because  they represent an asset with  considerable investment and mishandling  generates waste. Within the  types of  inventories, there are raw  materials takes, which  require greater control,  since   their proper planning  depends on the subsequent activities of the supply chain. By making a situational diagnosis at the company under study it was determined that there were economic losses due to inventory obsolescence  in the warehouse, delays in deliveries which led to fines and claims, demonstrating the need to apply an inventory  model that allows to reduce its costs and meet delivery deadlines. It starts by classifying inventories according to their level of importance with respect to the level of sales, in order to determine which of them are the most demanded by customers. The dispersion of the data is analyzed to form the database and look for outliers. The demand forecast was made with the Statistical Package for the Social Sciences (SPSS) and R Version 3.6.2, specifically with the package nnfor for neural networks. For the planning of the stock keeping unit (SKU) it's done with the models Silver Meal (SM), Wagner Whitin (WW) Fragmented Period Balance (BFP); determining an optimal result with a reduction of 57.89%, compared to the value of placing a monthly order during the period