巡游出租车运力规模动态调整回归树模型

针对巡游出租车运力规模调整指标体系的匮乏, 以宁波市出租车信息管理系统运营数据为基础, 梳理了巡游车运力规模调整的主要影响因素, 综合考虑各项运营指标, 构建了巡游车运力规模调整阈值回归树模型, 并采用方差分析的卡方自动交叉诊断器算法对模型进行标定. 最后, 耦合了万人拥有量、出租车在公交出行结构分担率、巡游车与网约车业务分担比以及乘客平均等候时间等指标在供需状态中的表征作用和重要性排序关系, 提出了巡游出租车运力规模动态调整机制及关键指标的阈值标准, 为城市巡游出租车运力规模调整提供理论依据和决策支持.

Regression tree model for dynamic adjustment of the scale of cruising taxicab capacity

In view of the lack of indicator system for the scale adjustment of the cruising taxicab capacity, this paper collects the operation datasets of taxicab from the Taxi Information Management System in Ningbo and analyzes the influential factors on cruising taxicab capacity. The regression tree model for the dynamic adjustment of the scale of cruising taxicab capacity is then established by considering the various operational indicators of the cruise taxicab. Next, the chi-squared automatic interaction detector is applied to calibrate the model. Finally, the dynamic adjustment mechanism of the scale of cruising taxicab capacity and the thresholds of the key indicators are proposed by coupling the representation functions on the balance of taxi supply and demand with the ordering relationship of importance of other indicators, which consist of the taxi ownership per person, the sharing ratio of taxi in public transport trip structure, the sharing ratio between cruising taxi and ride-hailing taxi, and the average waiting time of passengers, to provide theoretical basis and decision-making support for the scale adjustment of urban cruise taxi capacity.