The frequency and intensity of global disasters, including the COVID-19 pandemic, and natural disasters such as earthquakes, floods, and wildfires, are increasing, necessitating effective emergency logistics management. Climate change significantly contributes to these events, emphasizing the importance of limiting human and environmental impacts. The transportation sector, particularly the automobile industry, ranks second in global carbon emissions, highlighting the need to adopt electric vehicles (EVs) to reduce emissions and minimize the impact of climate change. However, this has led to an increase in demand for lithium-ion batteries. During emergencies, end-of-life (EOL) battery management through reverse logistics is essential because recycling EOL batteries can recover valuable raw materials, decrease landfill waste and costs, and support environmental sustainability. This study proposed a two-phase method for intelligent emergency EV battery reverse logistics management. The first phase employed machine learning to address unpredictable battery demands, whereas the second phase proposed a multi-objective model to minimize carbon emissions through efficient order allocation during uncertain emergencies. The model considers carbon emissions and defect rates as sources of uncertainty, current regulations, and customer environmental awareness. The model is solved using the weighted sum and ε-constraint methods, resulting in non-dominant solutions. The findings indicate that combining the selection of third-party reverse logistics providers (3PRLPs) with optimal order allocation for recycling old batteries during emergencies effectively minimizes environmental impacts and combats climate change.

中文翻译：

---

不确定下的电动汽车逆向物流：一种智能应急管理方法


全球灾害（包括 COVID-19 大流行）以及地震、洪水和野火等自然灾害的频率和强度正在增加，因此需要有效的应急物流管理。气候变化对这些事件有很大影响，凸显了限制对人类和环境影响的重要性。交通运输部门，尤其是汽车行业，在全球碳排放量中排名第二，这凸显了采用电动汽车 （EV） 来减少排放并最大限度地减少气候变化影响的必要性。然而，这导致对锂离子电池的需求增加。在紧急情况下，通过逆向物流进行报废 （EOL） 电池管理至关重要，因为回收 EOL 电池可以回收有价值的原材料，减少垃圾填埋场的浪费和成本，并支持环境可持续性。本研究提出了一种智能应急电动汽车电池逆向物流管理的两阶段方法。第一阶段采用机器学习来解决不可预测的电池需求，而第二阶段则提出了一个多目标模型，通过在不确定的紧急情况下进行有效的订单分配来最大限度地减少碳排放。该模型将碳排放和缺陷率视为不确定性、当前法规和客户环境意识的来源。该模型使用加权和ε约束方法求解，得到非主导解。研究结果表明，将第三方逆向物流提供商 （3PRLP） 的选择与紧急情况下回收旧电池的最佳订单分配相结合，可以有效地最大限度地减少对环境的影响并应对气候变化。