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 ›› 2026, Vol. 34 ›› Issue (11): 298-312.doi: 10.16381/j.cnki.issn1003-207x.2025.1921cstr: 32146.14.j.cnki.issn1003-207x.2025.1921

• • 上一篇    下一篇

基于AI预测性维护的供应链技术投入与融资策略研究

韩一萱1,2, 施先亮1, 张炜健1(), 刘帅3, 郑大昭2, 汪寿阳4,5   

  1. 1.北京交通大学经济管理学院,北京 100044
    2.香港理工大学工商管理学院,香港特别行政区 999077
    3.南京理工大学经济管理学院,江苏 南京 210094
    4.中国科学院大学经济与管理学院,北京 100190
    5.中国科学院数学与系统科学研究院,北京 100190
  • 收稿日期:2025-11-07 修回日期:2026-01-06 出版日期:2026-11-25 发布日期:2026-11-10
  • 通讯作者: 张炜健 E-mail:weijianzhang@bjtu.edu.cn
  • 基金资助:
    国家自然科学基金项目(72401283);国家自然科学基金项目(72501142)

Supply‑Chain Technology Investment and Financing Strategies under AI‑Based Predictive Maintenance

Yixuan Han1,2, Xianliang Shi1, Weijian Zhang1(), Shuai Liu3, T.C.E Cheng2, Shouyang Wang4,5   

  1. 1.Beijing Jiaotong University,School of Economics and Management,Beijing 100044,China
    2.Faculty of Business,The Hong Kong Polytechnic University,HongKong 999077,China
    3.Nanjing University of Science and Technology,School of Economics and Management,Nanjing 210094,China
    4.University of Chinese Academy of Sciences,School of Economics and Management,Beijing 100190,China
    5.Chinese Academy of Sciences,Academy of Mathematics and Systems Science,Beijing 100190,China
  • Received:2025-11-07 Revised:2026-01-06 Online:2026-11-25 Published:2026-11-10
  • Contact: Weijian Zhang E-mail:weijianzhang@bjtu.edu.cn

摘要:

人工智能是产业优化升级与生产力跃升的驱动力量,融资租赁作为制造业金融服务能够加速生产设备智能迭代。围绕资金受限制造商的设备采购与供应链是否引入基于AI的预测性维护决策,本文构建由设备供应商、制造商组成的两级博弈模型,决策技术投入与融资方式(银行借贷/融资租赁)的组合策略,并设计固定支付契约实现帕累托改进。研究结果表明:(1)相较银行借贷,融资租赁通过将利息成本内化并保留残值收益,削弱供应链利润外流,在不引入技术和供应商投入技术两类情形下,均能提高设备订购量和系统总产出,实现双方利润的帕累托改进。(2)基于AI的预测性维护能够通过降低非计划停机损耗并形成质量溢价,使得在提供租赁时供应商同时投入技术(情形Al)优于不投入;在仅采用银行借贷时,制造商采购设备同时投入技术(情形Amb)优于不投入。(3)银行利率是最优模式组合的关键影响因素。随着银行利率上升,最优模式由银行借贷下制造商自投技术(情形Amb)转向供应商投入技术并结合融资租赁(情形Al),且存在系统最优与供应商个体最优不一致的利率区间。针对该分歧,本文利用 Nash 议价确定最优固定支付,实现模式选择对齐与帕累托改进。研究结论对以融资租赁等金融服务推动 AI 技术投入落地、提升实体生产效率的产融结合实践具有参考价值。

关键词: AI预测性维护, 融资租赁, 供应链协调

Abstract:

Artificial intelligence (AI) is a major driver of industrial upgrading and productivity growth, while manufacturing-oriented financial services such as finance leasing can accelerate the intelligent upgrading of production equipment. Focusing on the equipment procurement decisions of a capital-constrained manufacturer and the supply chain’s adoption of AI-enabled predictive maintenance, a two-echelon game-theoretic model composed of an equipment supplier and a manufacturer is developed to analyze the joint strategy of technology investment and financing mode choice (bank borrowing versus finance leasing), and further designs a fixed-payment contract to achieve Pareto improvement. The results show that: (1) compared with bank borrowing, finance leasing internalizes interest costs and retains residual value within the supply chain, thereby mitigating profit leakage; both in the no-technology case and in the case where the supplier invests in the technology, finance leasing increases the equipment order quantity and total system output, leading to a Pareto improvement for both parties; (2) AI-enabled predictive maintenance reduces losses caused by unplanned downtime and generates a quality premium. As a result, when finance leasing is provided, the case in which the supplier also invests in the technology (Case Al) is superior to the case without technology investment; when only bank borrowing is adopted, the case in which the manufacturer invests in the technology while purchasing equipment (Case Amb) is superior to the case without technology investment; and (3) the bank interest rate is a key determinant of the optimal mode combination. As the bank interest rate rises, the optimal mode shifts from manufacturer-led technology investment under bank borrowing (Case Amb) to supplier-led technology investment combined with finance leasing (Case Al). Moreover, there exists an interest-rate interval in which the system-optimal mode is inconsistent with the supplier’s individually optimal mode. To address this divergence, Nash bargaining is employed to determine the optimal fixed payment, thereby aligning mode choice and achieving Pareto improvement. These findings provide useful insights for industry-finance integration practices that leverage financial services such as finance leasing to facilitate AI investment and adoption and enhance production efficiency in the real sector.

Key words: AI?based predictive maintenance, finance leasing, supply chain coordination

中图分类号: