Lin Marina, Schaposnik Laura P
Thomas Jefferson High School for Science and Technology, Alexandria, 22312, USA.
University of Illinois, Chicago, 60607, USA.
Sci Rep. 2025 Jul 1;15(1):21928. doi: 10.1038/s41598-025-08276-9.
According to the United States Environmental Protection Agency, transportation accounts for 28% of total U.S. emissions which is 8 billion tons of carbon dioxide, making it the largest contributor to the nation's greenhouse gas emissions. In an era where sustainability is becoming increasingly crucial, we introduce a novel Carbon-Aware Ant Colony System (CAACS) Algorithm that addresses the Generalized Traveling Salesman Problem while minimizing carbon emissions. We mathematically formulated the sustainable GTSP and developed an innovative approach that leverages the natural efficiency of ant colony pheromone trails to optimize routes, balancing both environmental and economic objectives. We discover new pathways achieving greater improvements in carbon emissions compared to the tradeoff in cost. Through our research, we developed several key concepts and correlations, including: a generalizable carbon emission heuristic that is adaptable to other carbon sources, the correlation that more ants improve solution quality and reduce runtime up to a threshold, and empirical proof of linear time complexity. The CAACS Algorithm identifies routes with carbon emissions less than or equal to the expected amount for 98% of instances in the benchmark datasets and in UPS Package Delivery we found a 0.02 % decrease in cost and 1.07 % decrease in carbon which can scale to millions of tons of carbon dioxide conserved in transportation. To the best of our knowledge, this is the first sustainable algorithm developed for the GTSP since the problem's introduction in 1969. By integrating sustainability into transportation models, the CAACS Algorithm is a powerful tool for real-world applications, including network design, delivery route planning, and commercial aircraft logistics. Our algorithm's unique bi-objective optimization represents a significant advancement in sustainable transportation solutions strategically balancing cost and carbon emissions to reduce energy consumption and promote environmental responsibility.
根据美国环境保护局的数据，交通运输在美国总排放量中占28%，即80亿吨二氧化碳，使其成为美国温室气体排放的最大贡献者。在一个可持续性变得越来越关键的时代，我们引入了一种新颖的碳感知蚁群系统（CAACS）算法，该算法在解决广义旅行商问题的同时，将碳排放降至最低。我们通过数学方法对可持续广义旅行商问题进行了公式化，并开发了一种创新方法，利用蚁群信息素踪迹的自然效率来优化路线，平衡环境和经济目标。我们发现了新的路径，与成本权衡相比，在碳排放方面有了更大的改善。通过我们的研究，我们开发了几个关键概念和相关性，包括：一种可推广到其他碳源的碳排放启发式方法；蚂蚁数量增加到一定阈值之前，能够提高解决方案质量并减少运行时间的相关性；以及线性时间复杂度的实证证明。CAACS算法在基准数据集中98%的实例中识别出碳排放小于或等于预期量的路线，在联合包裹服务公司的包裹递送中，我们发现成本降低了0.02%，碳排放量降低了1.07%，这可以扩展到交通运输中节省数百万吨二氧化碳。据我们所知，这是自1969年广义旅行商问题提出以来，为该问题开发的首个可持续算法。通过将可持续性整合到运输模型中，CAACS算法是一种适用于实际应用的强大工具，包括网络设计、配送路线规划和商用飞机物流。我们算法独特的双目标优化代表了可持续运输解决方案的重大进步，从战略上平衡成本和碳排放，以降低能源消耗并促进环境责任。
