关键词: 间歇性序列, 需求预测, 时间序列预测, 需求模式识别, 配件管理

Abstract: The demand for after-sales parts in large manufacturing enterprises is characterized by sparse distribution and high volatility, with high uncertainty in both demand frequency and quantity, presenting typical intermittent characteristics. However, in actual operation and maintenance, the demand for parts fluctuates greatly in terms of frequency and quantity, resulting in various demand patterns. The existing intermittent demand prediction mainly uses a single or static combination of fixed prediction models, which was difficult to fully explore the evolution law of demand sequences under different demand patterns, and the prediction accuracy and stability were hard to guarantee. To solve the above problems, an intermittent time series prediction method that adaptively matches demand patterns was proposed, which improved the prediction effect of intermittent sequences by dynamically identifying and matching demand patterns. The method included two stages. In the model training stage, firstly, according to the intermittent characteristics of the historical demand data of parts, it was divided into demand sequences and interval sequences, and the two types of sequences were clustered separately to capture the different demand and interval patterns corresponding to each type of sequence. Secondly, a prediction model library containing statistical analysis, shallow machine learning models, and deep learning models was established, and the prediction effects of each model on each demand pattern were tested to identify and mark the optimal prediction model for each type of demand pattern. In the prediction stage, the sequence to be predicted was divided into demand sequences and interval sequences, the demand pattern is identified and matched with the optimal prediction model, and the predicted values of demand and interval were combined to form the final prediction result. The experimental validation was carried out on the intermittent parts demand dataset of the American Automobile Company and the Royal Air Force, and the results showed that the proposed method could be applied to the historical data of parts with different demand patterns, and effectively improved the prediction accuracy by adaptively matching the demand patterns and the optimal prediction model, which provided an applicable demand prediction solution for the aftermarket services of large manufacturing enterprises.

Key words: intermittent sequence, demand forecasting, time series forecasting, demand pattern recognition, accessory management