Prediction of Power Material Demand with Small Samples Based on a Combination Model

In current, the high demands of power from various industries require power companies to manage power materials efficiently. Optimizing resource allocation is crucial for effective procurement planning and cost reduction. To this end, we propose, in this paper, a combination model to accurately predict the material demand in case of small samples provided by an electric power company. At first, three independent models were used to predict demand for four materials. Then the deviation comparison was carried out to determine the most reliable model. To further enhance the accuracy and dependability of material demand prediction, we combine the three independent models to generate a combination model, which outperforms the independent models. By using scientific material demand prediction, power enterprises can better understand the needs of the power market and make better procurement planning.