Abstract: Travel time reliability evaluation of adjacent bus roads contributes to making deep analysis on operational state of bus roads, finding weak roads and improving bus service level. The paper analyzes the distribution of adjacent bus roads travel time per mile. The Normal, Lognormal, Gamma and Weibull distributions are used to fit with bus roads based on sufficient field data. The lognormal is proved to be the fittest distribution. The paper also measures the threshold of bus roads travel time per mile under different service levels by percentile method. At the last section, the paper proposes a model for travel time reliability evaluation of adjacent bus roads based on service levels. The proposed method is evaluated by the operational data of bus line No. 337.

Key words: urban traffic, bus, adjacent bus roads, travel time reliability, lognormal distribution, bus service level

摘要： 研究常规公交微观区间运行时间可靠性,有助于从微观层面深入分析常规公交在各区间的运行状态稳定性，查找运行的薄弱段以提高公交服务水平.本文从研究常规公交微观区间运行时间的拟合分布入手，采用正态分布、对数正态分布、伽玛分布和威布尔分布，分别对不同性质城市道路微观区间单位距离运行时间进行分布拟合，样本分析结果表明，对数正态分布是最优拟合分布.根据分位数法确定了不同公交服务水平对应的微观区间单位距离运行时间阈值，在此基础上建立了基于可接受公交服务水平的常规公交微观区间运行时间可靠性模型，并对337路公交实测数据进行实例分析.

关键词: 城市交通, 常规公交, 微观区间, 运行时间可靠性, 对数正态分布, 公交服务水平