Variability in Regularity: Mining Temporal Mobility Patterns in London, Singapore and Beijing Using Smart-Card Data

Area of Science:

• Urban Dynamics and Mobility Studies
• Data Science and Urban Planning

Background:

• Understanding human mobility patterns is crucial for urban dynamics, city planning, and policymaking.
• Previous research identified universal regularities at aggregated scales, but finer scales reveal significant heterogeneity.

Purpose of the Study:

• To determine the temporal scales at which human mobility regularities are stable, explicable, and sustainable.
• To propose a measure of variability for assessing the stability of urban mobility regularities across different temporal scales.

Main Methods:

• Developed a basic measure of variability to assess the stability of mobility regularities.
• Analyzed one week of smart-card data from London, Singapore, and Beijing to compare urban mobility patterns.
• Examined temporal scales ranging from 1 minute to 24 hours to capture diurnal mobility cycles.

Main Results:

• Variations in mobility regularity scale non-linearly with temporal resolution, showing a dramatic increase in variability up to approximately 15 minutes.
• This suggests inherent limits for short-term mobility predictions.
• Beijing and Singapore exhibit higher regularity compared to London across all analyzed temporal scales.

Conclusions:

• Mobility regularity is scale-dependent, with significant increases in variability at finer temporal resolutions.
• The study provides a framework for comparative analysis of urban mobility data and highlights city-specific characteristics influencing regularity.
• Findings offer insights for policymakers to manage variability and improve urban travel experiences.