基于责任代币与 FedADMM 的低碳农产品供应链分布式优化配置方法

A Distributed Configuration Optimization Method for Low-Carbon Agri-Food Supply Chains Based on Responsibility Tokens and FedADMM

摘要

针对低碳农产品供应链中各主体因商业保密不愿公开成本、损耗率等内部参数，致使集中式优化难以适用的问题，提出一种基于责任代币与联邦交替方向乘子法 (FedADMM) 的分布式优化配置方法。在含碳限额与交易约束的配置模型中，引入时间与需求两类代币转移支付，将服务时间、累积成本与含损耗需求 3 类跨主体耦合松弛为经济违约惩罚；进而证明代币价格与 ADMM 对偶乘子等价，构造每轮仅交换聚合接口变量的 FedADMM 迭代算法，使成本等核心参数实现隐私保护。以苹果供应链为算例，FedADMM 求解结果与集中式方法一致，求解效率提升约一个数量级，参数暴露率由 100% 降至 0。敏感性分析表明，损耗率的总成本弹性比碳政策参数高约 2 个数量级，且在需求与碳价±50% 联合扰动下最优配置不变。该方法在保护主体隐私的同时获得全局最优解，为低碳农产品供应链分布式优化配置提供了有效途径。

Abstract

In low-carbon agri-food supply chains, centralized optimization is impractical since entities are often reluctant to disclose internal parameters such as costs and loss rates due to commercial confidentiality. To address this issue, a distributed configuration optimization method based on responsibility tokens and the federated alternating direction method of multipliers (FedADMM) is proposed. Two token transfer payments, namely time tokens and demand tokens, are introduced in a configuration model with carbon cap-and-trade constraints. Three categories of cross-entity couplings-service time, cumulative cost, and demand with loss-are thereby relaxed into economic penalties for contract violations. Token prices are then proved equivalent to the ADMM dual multipliers. Based on this, a FedADMM algorithm is built in which only aggregated interface variables are exchanged during each iteration, thereby preserving the privacy of core parameters such as costs. An apple supply chain is used as a case study. The proposed FedADMM obtains the same optimal solution as the centralized method, while improving computational efficiency by about an order of magnitude and reducing the parameter exposure rate from 100% to zero. Sensitivity analysis further shows that the elasticity of total cost with respect to the loss rate is about two orders of magnitude greater than that with respect to carbon-policy parameters. The optimal configuration remains unchanged under simultaneous ±50% perturbations in demand and carbon price. The proposed method reaches the global optimum while preserving the privacy of participating entities, offering an effective way for distributed configuration optimization in low-carbon agri-food supply chains.

关键词

供应链配置 / 农产品供应链 / 碳排放限额交易 / 分布式优化 / 责任代币 / FedADMM 算法

Key words

supply chain configuration / agri-food supply chain / carbon emission cap-and-trade / distributed optimization / responsibility tokens / FedADMM algorithm

参考文献

[1] Crippa M, Solazzo E, Guizzardi D, et al. Food systems are responsible for a third of global anthropogenic GHG emissions[J]. Nature Food, 2021, 2(3): 198-209.

[2] Graves S C, Willems S P. Optimizing the supply chain configuration for new products[J]. Management Science, 2005, 51(8): 1165-1180.

[3] 王琼, 张莉. 基于数字孪生的智能建造供应链系统构建与应用实践[J]. 建筑经济, 2026, 47(7): 60-68.

[4] Wang Qiong, Zhang Li. Construction and application practice of intelligent construction supply chain system based on digital twins[J]. Construction Economy, 2026, 47(7): 60-68.

[5] Chokri H, Nouaouri I, Allaoui H, et al. Agri-food supply chain network design: a comprehensive review and future research directions[J]. Computers & Industrial Engineering, 2025, 207: 111203.

[6] Camacho Á R E, Gholian-Jouybari F, Mosallanezhad B, et al. Designing a sustainable and resilient agri-food closed-loop supply chain network for porcine products[J]. Sustainable Futures, 2025, 9: 100591.

[7] Khazaei M, Mehrparvar M, Govindan K, et al. Circular supply chain design for biohydrogen recovery from perishable agri-food waste[J]. International Journal of Hydrogen Energy, 2026, 217: 153731.

[8] Nie D X, Li H T. Optimizing agri-food supply chain configuration for mitigating both supply- and demand-side risks[J]. Agriculture & Food Security, 2026, 15(1): 26.

[9] Li Z Y, Zhang C X. Designing a two-stage model for the resilient agri-food supply chain network under dynamic competition[J]. British Food Journal, 2024, 126(2): 662-681.

[10] Li H T, Li D, Jiang D L. Optimising the configuration of food supply chains[J]. International Journal of Production Research, 2021, 59(12): 3722-3746.

[11] 聂笃宪, 唐嘉燕, 赵金英, 等 . 基于人力资源约束和不确定需求的农产品供应链优化配置[J]. 工业工程, 2023, 26(2): 111-122.

[12] Nie Duxian, Tang Jiayan, Zhao Jinying, et al. An optimal configuration of agri-food supply chain based on labor resource constraints and uncertain demand[J]. Industrial Engineering Journal, 2023, 26(2): 111-122.

[13] 聂笃宪, 贝舒婕, 屈挺. 考虑碳交易与产品损失率的农产品供应链优化配置[J]. 供应链管理, 2023, 4(12): 34-50.

[14] Nie Duxian, Bei Shujie, Qu Ting. Optimal configuration of agri-food supply chain based on carbon trading and lost ratio[J]. Supply Chain Management, 2023, 4(12): 34-50.

[15] Das A, Pattnaik P K, Bandyopadhyay A, et al. Privacy preserving federated framework for autism spectrum disorder screening with ensemble stacking and deep learning approaches[J]. Discover Artificial Intelligence, 2026, 6(1): 426.

[16] Zhang B, Tan W J, Cai W T, et al. Forecasting with visibility using privacy preserving federated learning[C]// 2022 Winter Simulation Conference (WSC). Piscataway: IEEE, 2022: 2687-2698.

[17] Li J T, Cui T X, Yang K W, et al. Demand forecasting of E-commerce enterprises based on horizontal federated learning from the perspective of sustainable development[J]. Sustainability, 2021, 13(23): 13050.

[18] 马祁. 基于联邦学习的供应链节点信誉评估与资源配置方法[D]. 北京: 北京邮电大学, 2023.

[19] 崔天旭, 丁日佳, 华国伟, 等 . 联邦学习如何重塑电商供应链的数据共享方式?方法和案例[J/OL]. 中国管理科学, 1-18[2026-05-26]. https://doi.org/10.16381/j.cnki.issn1003-207x.2025.0264.

[20] Cui Tianxu, Ding Rijia, Hua Guowei, et al. How can federated learning reshape e-commerce supply chain data sharing? Method and case[J/OL]. Chinese Journal of Management Science, 1-18[2026-05-26]. https://doi.org/10.16381/j.cnki.issn1003-207x.2025.0264.

[21] 王健宗, 孔令炜, 黄章成, 等 . 联邦学习算法综述[J]. 大数据, 2020, 6(6): 64-82.

[22] Wang Jianzong, Kong Lingwei, Huang Zhangcheng, et al. Research review of federated learning algorithms[J]. Big Data Research, 2020, 6(6): 64-82.

[23] 周传鑫, 孙奕, 汪德刚, 等 . 联邦学习研究综述[J]. 网络与信息安全学报, 2021, 7(5): 77-92.

[24] Zhou Chuanxin, Sun Yi, Wang Degang, et al. Survey of federated learning research[J]. Chinese Journal of Network and Information Security, 2021, 7(5): 77-92.

[25] Boyd S, Parikh N, Chu E, et al. Distributed optimization and statistical learning via the alternating direction method of multipliers [J]. Foundations and Trends® in Machine Learning, 2011, 3(1): 1-122.

[26] Gholami A, Mirzazadeh A. An ADMM-based distributed optimization approach for make-to-order supply chains under Stackelberg game[J]. Computers & Industrial Engineering, 2022, 165: 107951.

[27] Zhang C L, Ahmad M, Wang Y Q. ADMM based privacy-preserving decentralized optimization[J]. IEEE Transactions on Information Forensics and Security, 2019, 14(3): 565-580.

[28] Wang Y, Yin W T, Zeng J S. Global convergence of ADMM in nonconvex nonsmooth optimization[J]. Journal of Scientific Computing, 2019, 78(1): 29-63.

[29] Porteus E L. Responsibility tokens in supply chain management[J]. Manufacturing & Service Operations Management, 2000, 2(2): 203-219.

基金资助

国家级大学生创新创业项目(202610564011)

华南农业大学校级项目(kcsz2024077)