Demand forecasting with artificial neural networks and implementation in the food industry

Companies should understand the events correctly and produce appropriate solutions within the framework of a good plan to protect and develop their current situation. For companies, the possibility of surviving is parallel with the accuracy of these predictions in increasing competition environment. According to this idea, demand forecast has vital importance for all companies. The purpose of the forecast is to predict situations companies may face in the future, using different data and techniques and take action in advance. Demand forecasting is to determine the demand level of a product or some products of a company for a specific time in the future. There are many methods used to forecast demand which are critical for companies. Artificial neural networks are a statistical methods which are efficiently used in solution of nonlinear problems and providing highly reliable results. Use of this method has increasingly spread in the demand forecasting. In this study, demand forecasting practice is made on delicatessen products from fresh food industry in FMCG industry using artificial neural network model. According to error test results, it has been observed that the forecasts of the model are reliable and consistent. Demand forecasts are prepared with also other methods and the results are compared. It's been shown that artificial neural networks is above existing methods. This study presents an example of demand forecast for other firms in the delicatessen fresh food industry.