Production planning is usually performed based on customer orders or demand forecasts. The aim of the study is to find a predictive analytics solution by modeling and simulation of different forecasting processes, i.e. forecast prediction and customer-provided forecasts. In addition, the paper analyzes the generated forecast data for which different forecast error measures are tested and discussed. In the case of customer-provided forecasts, a simple method for data generation is the application of independent forecast disturbances, i.e. the forecasts are stochastic and their updates are independent of the past forecast values.
To reach the goals, the researchers use AnyLogic as a predictive analytics solution. Different scenarios implemented in the simulation model are predefined by an input table and various parameters.