Impact of Bayesian Approach to Demand Management in Supply Chains for the Consumption of Dynamic Products

Abstract

Bayesian approach was applied to the management of the supply chain in a dynamic food product portfolio for a company in the retail sector. We propose a quasi-experimental method considering pre and posttest and a control group. The sample size of 93 products, out of a population of 120 products from two categories: classic sauces and gourmet sauces. R and Python programming languages were used and libraries for random sampling of the a priori distribution of the products to obtain posterior values area presented on the research. Forecast accuracy increased with the Bayesian approach by 10%. Likewise, it was possible to reduce the coverage inventory from 2 to 1.2 months and the discrepancy between the values of the Bayesian estimate with the traditional method was possible to reach a 5% error in the variation.