|
|
基于XGBoost模型产品订单量的预测
|
Abstract:
把握供应链需求，提高资源配置和利用效率对企业树立竞争优势具有重大的实际意义。通过特征分析提取了特征变量后，建立XGBoost模型，利用GridSearchCV技术对模型进行调参优化，最终得到模型的拟合优度为0.956，均方根误差RMSE为36.522。同时我们还建立了随机森林模型进行对比，结果表明XGBoost模型效果更理想，因此最终根据所构建的XGBoost模型实现对企业未来三月产品订单量的预测，为企业合理安排生产计划提供了一定的理论依据。
Grasping supply chain demand, improving resource allocation and utilization efficiency is of great practical significance for enterprises to establish competitive advantage. After extracting charac-teristic variables through feature analysis, XGBoost model is established, and the model is adjusted and optimized using GridSearchCV technology. Finally, the goodness of fit of the model is 0.956, and the Root-mean-square deviation RMSE is 36.522. At the same time, we also established a Random forest model for comparison. The results indicate that the XGBoost model has a more ideal effect. Therefore, based on the constructed XGBoost model, the prediction of the company’s product order volume in the next three months is ultimately achieved, providing a certain theoretical basis for the company to reasonably arrange production plans.
| [1] | Kim, W.J. (2018) Overseas Construction Order Forecasting Using Time Series Model. Korean Journal of Construction Engi-neering and Management, 19, 107-116. |
| [2] | Khashei, M. and Bijari, M. (2010) An Artificial Neural Network (p, d, q) Model for Timeseries Forecasting. Expert Systems with Applications, 37, 479-489. https://doi.org/10.1016/j.eswa.2009.05.044 |
| [3] | Voronin, S. and Partanen, J. (2014) Forecasting Electricity Price and Demand Using a Hybrid Approach Based on Wavelet Transform, ARIMA and Neural Networks. International Journal of En-ergy Research, 38, 626-637.
https://doi.org/10.1002/er.3067 |
| [4] | 王斌, 杨抒, 贾清, 赵毅, 王业. ARIMA模型在电商平台新疆灰枣订单预测中的应用研究[J]. 福建电脑, 2019, 35(11): 5-8. https://doi.org/10.16707/j.cnki.fjpc.2019.11.002 |
| [5] | 谭祖健, 官丹萍, 莫愁, 丁振伟, 李晓霞. 基于非平稳时间序列分析的发动机订单预测模型[J]. 中国新技术新产品, 2022(4): 16-19. https://doi.org/10.13612/j.cnki.cntp.2022.04.005 |
| [6] | 施佳. 基于ARIMA-BP组合模型的某餐饮O2O企业订单预测研究[D]: [硕士学位论文]. 北京: 北京交通大学, 2018. |
| [7] | 欧红森, 姚玉南, 冯皓, 何溢. 基于GA-灰色BP神经网络船舶备件需求预测模型[J]. 中国修船, 2023, 36(2): 41-46. https://doi.org/10.13352/j.issn.1001-8328.2023.02.012 |
| [8] | 李秀丽, 孙国华. 基于机器学习的产品需求量预测方法研究[J]. 科技视界, 2020, 24(2): 98-99. |
| [9] | Zhao, Y., Qi, Y. and Dong, J. (2019) Product Demand Forecasting Based on Online Customer Reviews: A deep Learning Approach. Journal of Business Research, 96, 365-377. |
| [10] | 刘建华, 李军. XGBoost算法原理及应用[J]. 计算机科学, 2017, 44(12): 1-7. |
| [11] | 周志华, 吴恩达, 等. XGBoost: 基于提升树的高效机器学习算法[J]. 计算机学报, 2016, 39(3): 530-543. |
| [12] | 曹雷, 尚维, 谢士尧, 王向. 基于AGNN舆情指数网络的价格指数预测研究[J]. 管理学报, 2023, 20(3): 411-421+431. |
