主管:中国科学院
主办:中国优选法统筹法与经济数学研究会
   中国科学院科技战略咨询研究院

中国管理科学 ›› 2025, Vol. 33 ›› Issue (12): 200-213.doi: 10.16381/j.cnki.issn1003-207x.2023.1836cstr: 32146.14.j.cnki.issn1003-207x.2023.1836

• • 上一篇    下一篇

基于数据挖掘的电力装备制造企业多价值链协同下的成本预测研究

许晓敏(), 郑世鹏, 王之怡, 姚润坤, 关泺允   

  1. 华北电力大学经济与管理学院,北京 102206
  • 收稿日期:2023-11-03 修回日期:2024-10-17 出版日期:2025-12-25 发布日期:2025-12-25
  • 通讯作者: 许晓敏 E-mail:xuxiaomin0701@126.com
  • 基金资助:
    国家重点研发计划项目(2020YFB1707802);中国科协青年人才托举工程项目(YESS20220084)

Research on Cost Prediction of Electric Power Equipment Manufacturing Enterprises under Multi Value Chain Collaboration Based on Data Mining

Xiaomin Xu(), Shipeng Zheng, Zhiyi Wang, Runkun Yao, Luoyun Guan   

  1. School of Economics and Management,North China Electric Power University,Beijing 102206,China
  • Received:2023-11-03 Revised:2024-10-17 Online:2025-12-25 Published:2025-12-25
  • Contact: Xiaomin Xu E-mail:xuxiaomin0701@126.com

摘要:

在电力装备制造企业多价值链协同的背景下,企业成本预测受供应链、生产链、营销链、服务链的多重影响。为提升电力设备制造企业成本预测精度和成本管理水平,本文构建了一种基于数据挖掘技术的萤火虫扰动(firefly algorithm, FA)和麻雀搜索算法(sparrow search algorithm, SSA)组合优化BP神经网络(FA-SSA-BP)的成本预测模型。首先,利用网络数据挖掘技术,构建电力装备制造企业多价值链协同下的经营成本影响因素库。其次,运用Pearson相关系数和灰色关联方法对影响因素库进行筛选,确定关键因素。然后,构建FA-SSA-BP成本预测模型,其中,FA-SSA强化了全局搜索能力,避免过早陷入局部最优,提升了收敛精度。最后,利用Q电力设备制造企业主营业务环网柜的相关数据进行实证分析,并对关键因素进行敏感性分析,提出电力装备制造企业多价值链协同下的成本管理建议。

关键词: 多价值链协同, FA-SSA-BP, 成本预测, 数据挖掘, 电力装备制造企业

Abstract:

In the context of multi value chain collaboration in power equipment manufacturing enterprises, enterprise cost prediction is influenced by multiple factors such as supply chain, production chain, marketing chain, and service chain. In order to improve the accuracy of cost prediction for power equipment manufacturing enterprises and improve the level of cost management, a cost prediction model is constructed based on data mining technology, which combines firefly algorithm (FA) and sparrow search algorithm (SSA) optimization BP neural network (FA-SSA-BP). Firstly, using network data mining technology, a database of influencing factors for multi value chain collaboration in power equipment manufacturing enterprises is constructed; Secondly, the Pearson correlation coefficient and grey relational analysis (GRA) method are used to screen the influencing factor library and determine key factors; Then, a FA-SSA-BP cost prediction model is constructed, in which FA-SSA strengthened the global search ability, avoided premature falling into local optima, and improved convergence accuracy. Subsequently, based on the relevant data of the main business Ring Main Unit of Q power equipment manufacturing enterprise, the prediction model constructed in this paper is used for cost prediction and error analysis, and the prediction effect is compared with other optimization models. The results indicate that the model proposed in this paper significantly reduces prediction errors and effectively improves the accuracy of cost prediction compared to the comparative model. Then, sensitivity analysis is conducted on key factors, pointing out the impact of different influencing factors on enterprise costs. Finally, based on the above research, targeted cost management suggestions are proposed for power equipment manufacturing enterprises under multi value chain collaboration. The cost prediction model constructed in this paper and the proposed management suggestions lay a solid foundation for cost management in power equipment manufacturing enterprises under multi value chain collaboration.

Key words: multi value chain collaboration, FA-SSA-BP, cost prediction, data mining, power equipment manufacturing enterprise

中图分类号: