Hierarchical Agglomerative Clustering using GPS Data for Defining Level of Service Criteria of Urban Streets

摘要: This paper describes how speed ranges for Level of Service (LOS) are presently not well defined highly heterogeneous traffic flow on urban streets in developing countries. In this respect, a study was carried out the city Mumbai, India by authors. The uses average travel segments as measure effectiveness, which case has been derived from second data collected using Global Positioning System (GPS) receiver fitted mobile vehicle. Applied technique, Hierarchical Agglomerative Clustering (HAC) is way to investigate grouping creating cluster tree called dendogram and cutting it at desired level, then each connected component forms cluster. Hence possibility obtaining new LOS criteria emerges. Applying methodology found that street speed-ranges valid Indian conditions different corresponding values specified Highway Capacity Manual 2000 (HCM 2000). It proposed present work quantitative method measuring can further be extended other methods like k-means clustering fuzzy c-means clustering.