DEMAND FORECASTING BY ECONOMETRIC MODELS SUPPORTING OPERATIONAL OPTIMIZATION AND EMPLOYABILITY IN TRANSPORT

Abstract

Through the use of suitable methods for analyzing future demand, it is possible to quickly deal with fluctuations originating from external factors of various characteristics that may interfere with the planning of production systems, which is essential for sizing the availability of assets, inputs and, above all, human resources. It is in this bias that this article is framed. The objective of this article is to present the potential and simplicity of using stochastic methods for forecasting demand using historical series to assess the future employability of transport, through a case study. Data from the primary base of the new CAGED/CNT are used, allowing prospecting for future data in applied and descriptive research for the interpretation of the operational scenario and human resources of cargo transport. For that, three stochastic methods are used to generate the values of the random and systematic components of the projected demand. After analyzing the random component and the bias ratio, the best Holt method is identified as the one that fits in the range with the smallest amplitude for any measure of the difference between forecast and actual demand. By prospecting ten periods ahead, the idea of optimizing the infrastructures of the modes of transport was composed to minimize layoffs, in addition to greater intensity of multimodality operations. With the study it is possible to perceive the importance and potentiality of the use of stochastic methods for forecasting demand to improve operational results, in addition to improving the efficiency of the application of human resources in the transport of cargo, which can impact on less variability in balances of jobs in the various load modes.