[Paper Review] Portraying ride-hailing mobility using multi-day trip order data: A case study of Beijing, China

Zhengbing He|arXiv (Cornell University)|Jun 23, 2020

Transportation and Mobility Innovations45 references44 citations

TL;DR

This study leverages multi-day ride-hailing trip order data from Beijing to analyze spatiotemporal mobility patterns from both regional and driver perspectives. It reveals two distinct driver types—those with activity spaces linearly correlated to working hours and those restricted to limited areas—while identifying power-law distributions in trip duration and displacement, and a strong diurnal rhythm in driver activity, significantly advancing understanding of on-demand mobility and informing service management and demand prediction.

ABSTRACT

As a newly-emerging travel mode in the era of mobile internet, ride-hailing that connects passengers with private-car drivers via an online platform has been very popular all over the world. Although it attracts much attention in both practice and theory, the understanding of ride-hailing is still very limited largely because of the lack of related data. For the first time, this paper introduces ride-hailing drivers' multi-day trip order data and portrays ride-hailing mobility in Beijing, China, from the regional and driver's perspectives. The analyses from the regional perspective help understand the spatiotemporal flowing of the ride-hailing demand, and those from the driver's perspective characterize the ride-hailing drivers' preferences in providing ride-hailing services. A series of findings are obtained, such as the observation of the spatiotemporal rhythm of a city in using ride-hailing services and two categories of ride-hailing drivers in terms of the correlation between the activity space and working time. Those findings contribute to the understanding of ride-hailing activities, the prediction of ride-hailing demand, the modeling of ride-hailing drivers' preferences, and the management of ride-hailing services.

Motivation & Objective

• To understand the spatiotemporal dynamics of ride-hailing demand in Beijing using real multi-day trip order data.
• To characterize ride-hailing drivers’ behavioral preferences in service provision, particularly regarding working hours and spatial activity.
• To identify structural patterns in ride-hailing mobility that differentiate it from traditional taxi services.
• To provide data-driven insights for transportation planners and ride-hailing platform operators on service supply and demand forecasting.
• To establish a foundation for modeling driver choice behavior and predicting regional demand with high temporal and spatial precision.

Proposed method

• Utilizes a comprehensive dataset of multi-day ride-hailing trip orders from a major transportation network company (TNC) in Beijing.
• Applies spatial clustering to identify drivers' origin and destination patterns and define their activity spaces.
• Employs temporal analysis to examine the distribution of working hours across weekdays and weekends.
• Conducts statistical analysis on trip duration and displacement distributions, identifying power-law tail behavior.
• Uses grid-based spatial analysis (1 km resolution) to map regional demand intensity and temporal patterns.
• Combines temporal and spatial characteristics to classify drivers into two distinct behavioral categories based on correlation between working time and activity space.

Experimental results

Research questions

• RQ1 How does ride-hailing demand vary across different regions and times of day in Beijing?
• RQ2 What are the dominant temporal patterns in ride-hailing driver activity, and how do they differ between weekdays and weekends?
• RQ3 How do ride-hailing drivers spatially distribute their service areas, and what patterns emerge in their activity spaces?
• RQ4 What are the statistical distributions of ride-hailing trip durations and displacements, and do they exhibit power-law behavior?
• RQ5 Are there distinct behavioral categories of ride-hailing drivers based on the relationship between their working hours and spatial activity?

Key findings

• The period [12:00, 18:00] contains the highest proportion of frequently working drivers (16.5% on weekdays and 14.1% on weekends), indicating peak service availability.
• 18.7% of frequently working drivers choose to work approximately 6 hours per day, suggesting a preference for short, flexible shifts compared to traditional full-time work.
• Trip duration and displacement distributions exhibit power-law tails with exponents of -4.27 and -4.52, respectively, indicating occasional long trips despite most trips being short and local.
• Only 8% of drivers operate in areas larger than Beijing’s central urban region, while 19% are active in very small spatial zones, highlighting strong spatial concentration.
• 28% of drivers provide services across the entire city, but most drivers restrict their operations to a limited spatial scope, indicating strong spatial preferences.
• Two distinct driver categories emerge: one with activity space positively correlated to working time, and another who operate within a fixed, confined area, confirming driver selection behavior as a key differentiator from taxi drivers.