题目：Research on the application of deep learning in low-carbon supply chain management（深度学习在低碳供应链管理中的应用研究）
作者： Zhao Tian and Lou Junting
摘要：Based on panel data from 204 prefecture-level cities in China spanning from 2010 to 2019, this study examines the low-carbon city pilot program as a quasi-natural experiment. The propensity score matching-double difference model (PSM-DID) is utilized to analyze the effects of the low-carbon city pilot policy on carbon emission intensity. The research findings found: Firstly, the low-carbon city pilot policy has significantly reduced the intensity of urban carbon emissions. Secondly, the primary factors contributing to the reduction in urban carbon emissions intensity due to the low-carbon city pilot policy include adjustments in industrial structure, energy consumption control, and advancements in green technology innovation. Thirdly, heterogeneity analysis indicates that the impacts of the low-carbon city pilot policy differ across various regions. Resource cities, eastern cities, as well as cities in the southeast and northwest regions, experience significant effects.
本研究提出了一种基于深度学习的框架，旨在提升 低碳供应链管理（LCSCM）的效率与可持续性。首先，构建了多尺度时间序列分解长短期记忆网络（MS-TDLSTM）模型，该模型融合经验模态分解（EMD）与注意力机制，能够精准捕捉碳排放数据的多尺度特征。其次，设计了基于深度强化学习（DRL）的多目标优化模型。通过软约束多目标强化学习方法，将预测与优化过程整合为统一系统，实现了低碳供应链管理的智能决策。
原文： Zhao T , Lou J .Research on the application of deep learning in low-carbon supply chain management[J].International Journal of Low Carbon Technologies, 2025, 20(000):209–216.
链接：https://academic.oup.com/ijlct/article/doi/10.1093/ijlct/ctae290/7990584
当前位置：