Demand forecasting methods: a review of the literature

Abstract

Demand forecasting is a business management methodology for estimating the future value of a large interest. Carrying out demand forecasts means reconfiguring behavioral patterns in historical series and predicting future behavior or, furthermore, identifying causal factors that affect behavior and extrapolate it. The objective of this article is to carry out a review of the literature on two demand forecasting methods with the purpose of bringing together the methods and models available on two concepts currently used in the administration of companies related to the consumption and production of products and services. The methodology used is a review of the literature with a qualitative approach, with the purpose of giving a general descriptive view of two dominant methods used in forecasting demand. The mapping of the literature was carried out to identify the state of science through the scientific production available in the Scopus and Google Scholar data banks. Qualitative methods and causes are more associated with forecasts of medium and long terms. In the analysis of time series, as the methods, two different types of means and exponential smoothing are indicated as the most suitable for short-term forecasts. A resource used in various realities is the construction of its own demand forecasting model, or one that uses techniques, aspects, concepts and characteristics of different methods and models. It is essential to monitor the adopted model, keep the field and forecast data under control and, if there are any deviations, correct the model.