【摘要】 由于现有微循环公交研究大多忽视老龄群体的出行需求与偏好。本文利用城市兴趣点数据、公交刷卡数据和问卷调查数据所得的出行需求和出行意愿优先级，改进AP聚类算法，解决老龄化社区的微循环公交站点选址问题。基于选定的站点，以车辆运营成本和乘客在途成本最小为目标建立老龄化社区微循环公交动态线路规划模型并使用遗传算法求解。以北京市东城区老龄化程度为33.2%社区进行实例研究，结果表明：经改进AP聚类算法所得的公交站点可覆盖老龄群体的日常出行需求。在此基础上，动态线路规划模型得出的最优解显示，线路总运营成本为23.95元，线路总行驶时间为0.30 h,乘客人均在途成本为1.24元，结果验证了模型和算法的可行性。更多还原

【Abstract】 Most existing studies on micro-circulation public transportation have ignored the travel needs and preferences of the elderly. This paper uses the travel needs and travel intention priorities obtained from urban point of interest data, bus card swiping data and questionnaire survey data to improve the AP clustering algorithm to solve the site selection problem of aging communities. Based on the goal of minimizing vehicle operating costs and passenger in-transit costs, a dynamic route planning model for micro-circulation public transportation in aging communities is established and solved using a genetic algorithm. A case study is conducted on a community in Dongcheng District, Beijing, where the aging rate is 33.2%. The results show that the bus stops obtained by the improved AP clustering algorithm can cover the daily travel needs of the elderly. Based on the site selection of the stops, the total operating cost of the route after the model optimization is 23.95 yuan, the total travel time of the route is 0.30 hours, and the average in-transit cost per passenger is 1.24 yuan. The results of the study verified the feasibility of the model and algorithm.更多还原