Previsão de demanda e gestão de estoque aplicada a uma indústria pet food

Abstract

The integration of sectors in a growing company can be considered one of the main factors that make it difficult to use information, harming its performance in a competitive market. As a solution, management tools such as demand forecasting can help in better decision making in relation to stock, purchases and production planning. To this end, this case study aimed to identify the demand forecasting method with the least error based on a time series and such as the Economic Order Quantity, Safety Stock and Maximum Inventory Level. The application was carried out in a small pet food industry, considered one of the hottest segments in the country, by collecting data on the product with the highest revenue of the company. Using comparative methods, the SARIMA (96,0,22)(0,0,2,30) model presented an RMSE error 32.70% lower compared to the Holt model, and 27.04% lower compared to the HoltWinters, being the most suitable model for the application of the Economic Order Quantity (Q) model for this product. Thus, raising the costs involved in the stock and calculating the cost of ordering and storage, it was possible to obtain a value of Q 29,645kg, a Safety Stock of 19,300kg and an average of Maximum Inventory Level of 58,038kg considering the 48 expected deadlines by the demand forecast model. It was considered that the study was prepared in a planned way to propose a proposal for better results for the management company.