Detecting Taxi Travel Patterns using GPS Trajectory Data: A Case Study of Beijing

Abstract

GPS trajectory is a valuable source to understand the operational status of taxicabs and identify the traffic demand and congestions. This study attempts to use 24-hour taxi trajectory data to investigate the attributes of taxicabs such as the distance of occupied distance, number of active taxicabs in different hours, average trip speed in different hour, coverage area of a taxicab, average radius of a taxicab, occupied rate and service times. The results show that the highest speed of taxicabs occur in the 3:00 am when there is the smallest number of active taxicabs running on the road. Moreover, the average occupied rate is 0.59 and the average service times are 19.8 in a day. Finally, a latent class analysis model is used to make the segment of taxicabs by their attributes. Four operational patterns have been found including ‘downtown preference type’, ‘long-distance preference type’, ‘suburbs preference type’ and ‘free preference type’. This study can shed light on understanding the operational status of taxicabs and gives suggestions for operators and passengers for better managing and using taxicabs.

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 — 21 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.