Simulating two-phase taxi service process by random walk theory.

Area of Science:

• Complex Systems Science
• Transportation Science
• Data-Driven Modeling

Background:

• Existing data-driven methods for city taxi services lack comprehensive models.
• Current mathematical models fail to capture the full taxi service process, including distinct pick-up and drop-off phases.

Purpose of the Study:

• To analyze taxi service data to understand distinct temporal and spatial features of pick-up and drop-off phases.
• To develop independent mathematical models for simulating taxi service mechanisms.
• To provide a framework for better understanding taxi service dynamics.

Main Methods:

• Analysis of large-scale taxi service data from a major Chinese city.
• Correlation analysis to determine interdependencies between service phases.
• Development of two independent random walk models based on the Langevin equation.

Main Results:

• Taxi service processes exhibit different temporal and spatial characteristics during the on-load (pick-up) and off-load (drop-off) phases.
• A lack of significant correlation was found between the on-load and off-load phases.
• The proposed models successfully describe the observed temporal and spatial features.

Conclusions:

• The on-load and off-load phases of taxi services operate independently with distinct characteristics.
• Random walk models based on the Langevin equation provide a robust framework for simulating and understanding taxi service mechanisms.
• This research offers a novel mathematical approach to urban transportation system analysis.