Methods of forecasting stocks of construction materials during deliveries

Introduction. Dynamic development of retail trade in construction materials increases the requirements for timely delivery of goods to store warehouses. Well-established classical algorithms are focused on calculating the target inventory by taking into account the sales history, which characterizes real demand, because it is subject to distortions caused by the influence of marketing campaigns, stock shortages and abnormal sales. Under such conditions, it is incorrect to predict inventories using the classical algorithm. The evolution of forecasting methods is characterized by a shift in emphasis from demand for goods to inventory management. For this reason, it is necessary to develop the practice of modelling orders to suppliers of construction materials. In turn, there is a problem of forecasting the supply of stocks of construction materials. The purpose of the paper is to assess the capabilities of existing methods of forecasting stocks of construction materials of a particular group during deliveries. Research objectives: to analyze the possibilities of existing forecasting methods for the task of inventory management; to carry out the necessary statistical calculations for forecasting inventories.Materials and methods. Methods of theoretical analysis of scientific literature, statistical data analysis and comparative analysis, method of calculating the root mean square error of modelling RMSE, Holt method and simulation modelling were used for the research tasks.Results. Based on the root mean square error of the RMSE modeling, the size of the error is established for each of the analyzed inventory forecasting methods.Conclusions. Based on the calculations, it is determined that the most optimal method for forecasting inventories of construction materials is the method of simulation modelling, since it allows forecasting with the smallest degree of error.

construction logistics, sliding curve, simulation modelling, construction business, forecasting, Holt method