军事知识服务平台
基于深度强化学习的供应链碳金融减排效果
The Emission Reduction Effectiveness of Supply Chain Carbon Finance Based on Deep Reinforcement Learning
摘要 本研究构建了一个基于深度强化学习的动态碳减排融资框架,旨在探讨在有无政府补贴情境下供应链碳金融如何影响中小运输企业碳减排决策.研究结果表明:首先,供应链碳金融模式可以推动所有中小运输企业提升碳减排量,达到较高的碳减排均衡状态;尤其是在政府补贴政策后,这些企业的碳减排均衡水平将显著高于没有补贴时的水平.其次,当第三方碳信息平台更关注中小运输企业的碳减排行为,而非仅是碳减排程度时,能更有效地激励中小运输企业碳减排.最后,本文将深度强化学习应用于供应链系统仿真,并结合实际企业数据进行了细致分析,开辟了评估政策对中小运输企业微观行为影响的新视角.该动态碳减排融资框架为企业决策提供了强有力的支持,使企业得以借助深度强化学习方法,有效适应市场变动并灵活调整其碳减排策略,展现了其在实际应用中的巨大潜力和价值.
Abstract:
This study constructs a dynamic carbon emission reduction financing framework based on deep reinforcement learning(DRL),aimed at exploring how supply chain carbon finance(SCCF)influences the carbon reduction decisions of small and medium-sized transportation enterprises(SMTEs)under scenarios with and without government subsidies.The findings reveal that:First,the SCCF model can motivate all SMTEs to enhance their carbon reduction efforts,achieving a higher equilibrium state of carbon reduction;notably,the equilibrium level of carbon reduction for these enterprises significantly surpasses that of the non-subsidy scenario upon the implementation of government subsidies.Second,when third-party carbon information platforms focus more on the carbon reduction actions rather than the degree of reduction actions of SMTEs,which can more effectively motivate these enterprises to undertake reduction measures.Additionally,this paper applies DRL to supply chain system simulation for and conducts a detailed analysis with real enterprise data for detailed simulation analysis,thus providing a new perspective for assessing the impact of policies on the micro-behaviors of SMTEs.This dynamic carbon reduction financing framework offers robust support for enterprise decision-making,enabling businesses to effectively adapt to market changes and adjust their carbon reduction strategies through DRL,showcasing its significant potential and value in practical applications.
刊 名 系统工程 ISTICPKU
年,卷(期): 2026, 44(2)
分类号: F272
机标分类号: F512.6 F272.3 F832.33
在线出版日期: 2026年6月5日
基金项目: JCZRQNB202600202:湖北省自科基金项目,