Optimizing Simple Exponential Smoothing for Time Series Forecasting in Supply Chain Management

Abstract

This paper deals with optimizing Simple Exponential Smoothing for time series forecasting in supply chain management, particularly in the transport and automotive sectors. This paper attempts to enhance the accuracy of the forecast by estimating an optimal smoothing constant α with the Mean Squared Error as the objective function. This optimization exercise will be done using MATLAB's fminsearch function. Indeed, results realize substantial improvements in the accuracy of the forecast, validated using different error metrics and graphical representations.