Description
The need to establish effective demand management is a challenging activity for organizations and dealing with market uncertainties is a key activity to ensure the growth and competitiveness of a company. In an era where it is possible to store large amounts of data, using advanced analytical techniques can bring useful information to decision makers. Thus, the present work was developed in a medium-sized industry, inserted in the asphalt paving market productive chain, aiming to use a demand forecasting method, using a predictive model with Machine Learning. Using the Law and Kelton method for modeling construction, a study was conducted and it was decided to use the R software and apply the Prophet algorithm to analyze the product time series and build a model for validation. Then the model was built, adjusted to simulate the actual demand of the subsequent months. The results were evaluated and interpreted, obtaining a maximum error of 2.8% and SEM of -7%. The use of Machine Learning for demand forecasting was satisfactory and can be considered as an effective tool for decision making. A critical analysis of the method and suggestions for future research concludes this study.
