In the context of ‘carbon peak’ and ‘carbon neutrality’, coordinating individual travel demand through multi-modal transportation and guiding travelers towards new shared public transportation (PT) modes is increasingly important. In this paper, we analyze the competitive and cooperative relationship between online car-hailing (OCH) services and metro systems in Nanning, China, and conduct aquestionnaire survey among different types of OCH users. A mixed choice model that considers psychological latent variables is constructed to investigate OCH users’ attitudes and cognitions toward customized buses (CBs). An improved adaptive Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm is proposed to identify potential carpooling station sets, and a hybrid genetic-ant colony algorithm (GACA) is designed to solve bi-level programming model for CB line optimization. Case study results indicate an 83.8% overall transfer rate from OCH users to CBs, with the optimized scheme achieving a 69.68% reduction in carbon emissions.

中文翻译：

---

地铁与网约车合作竞争关系下考虑用户换乘意愿的定制公交线路优化


在“碳达峰”和“碳中和”的背景下，通过多式联运协调个人出行需求，引导出行者采用新型共享公共交通（PT）模式变得越来越重要。本文分析了南宁市网约车服务与地铁系统的竞争与合作关系，并对不同类型的网约车用户进行了问卷调查。构建考虑心理潜变量的混合选择模型来调查OCH用户对定制公交车（CB）的态度和认知。提出了一种改进的自适应噪声应用密度空间聚类（DBSCAN）算法来识别潜在的拼车站集，并设计了一种混合遗传-蚁群算法（GACA）来求解CB线路优化的双层规划模型。案例研究结果表明，从OCH用户到CB的总体转移率为83.8%，优化方案实现碳排放减少69.68%。