The Cluster of Vehicle Temporal Travel Behavior Based on License Plate Recognition Data

Abstract

The analysis for vehicle travel characteristics has always been of great interest to transport authorities, since it has a significant impact on logistic and operational decisions. Moreover, vehicle clustering can improve market research through more targeted access to groups of interest and facilitate planning through better survey design. This paper directly clustered 392,856 vehicles in a week using k-means clustering algorithm based on license plate recognition (LPR) data obtained in Shenzhen, China. First, several corresponding temporal variables are applied in weekdays and weekends respectively to identify homogeneous clusters. Second, Davies Bouldin index (DBI) and silhouette coefficient (SC) are utilized to find the optimal cluster number. Finally, seven groups in weekdays and three in weekends are classified. Meanwhile, detailed analysis of the characteristics for each group in terms of temporal changes in cluster characteristics is presented.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 6 citations exist upstream.