Divisive Analysis (DIANA) of hierarchical clustering and GPS data for level of service criteria of urban streets

Abstract

Level of Service (LOS) for heterogeneous traffic flow on urban streets is not well defined in Indian context. Hence in this study an attempt is taken to classify urban road networks into number of street classes and average travel speeds on street segments into LOS categories. Divisive Analysis (DIANA) Clustering is used for such classification of large amount of speed data collected using GPS receiver. DIANA algorithm and silhouette validation parameter are used to classify Free Flow Speeds (FFS) into optimal number of classes and the same algorithm is applied on speed data to determine ranges of different LOS categories. Speed ranges for LOS categories (A–F) expressed in percentage of FFS are found to be 90, 70, 50, 40, 25 and 20–25 respectively in the present study. On the other hand, in HCM (2000) it has been mentioned these values are 85 and above, 67–85, 50–67, 40–50, 30–40 and 30 and less percent respectively.