【摘要】 自行车骑行关键路径识别能够为非机动车道骑行空间品质改善优先级提供依据。创新定义自行车骑行关键路径评价指标，考虑到共享单车骑行路径与自行车骑行路径具有一致性，依托共享单车骑行轨迹数据，提出基于频繁项集自组合的关联规则算法识别自行车关键路径。通过ArcGIS平台将共享单车GPS轨迹数据与非机动车路段ID匹配，借助频繁项集自组合的Apriori算法构建自行车骑行关键路径算法。选择厦门市部分区域共享单车GPS轨迹数据进行算例分析，结果表明，提出的方法相较于原始算法计算速度提高4.43倍，能够识别早高峰自行车骑行关键路径，并对关键路径进行排序，可用于非机动车道管理改善优先级的确定。更多还原

【Abstract】 The identification of critical paths for bicycle riding can provide a basis for improving the quality of cycling space in non-motorized lanes and determine the priority of lane improvement.Redefining the evaluation index of the critical path of bicycle riding.Considering the consistency between the shared bicycle riding path and the bicycle riding path,the paper relies on the riding trajectory data of the shared bicycle,and then proposes an association rule algorithm based on frequent itemsets self-combination,used to identify bicycle critical paths.Through the ArcGIS platform,the shared bicycle GPS trajectory data is matched with the non-motor vehicle road segment ID data.The Apriori algorithm of frequent itemsets self-combination to construct a bicycle cycling critical path algorithm.Selecting the GPS trajectory data of shared bicycles in some areas of Xiamen City for example analysis,the results show that the method proposed in this paper is 4.43 times faster than the original algorithm.The method proposed in this paper can be used for prioritization of future bike lane facility management improvements.更多还原