Research in Transportation Business & Management'te Makalemiz Yayımlandı - 2026.09.07
The new article co-authored by CEDS Lab coordinators Gürkan Çelik and Murat Alper Başaran, which investigates pre-ride walking behavior in dockless shared mobility systems, has been published.
This study addressed pre-ride walking behavior, which is the walk a user takes to reach a shared vehicle. Instead of relying on surveys or broad assumptions, a massive dataset of 59 million location telemetry logs and 1 million trip records across 10 cities was analyzed. By developing a precise matching algorithm, the data revealed that 31.5% of all trips started with a walk of just 1 minute or less. Furthermore, it was determined that steep elevations, cool weather, and even the number of service providers operating in a city fundamentally altered a user's walking behavior.
An Artificial Neural Network model was developed to predict these walking durations under specific constraints. This behavioral insight served as a computational engine for advanced, dynamic vehicle positioning strategies, allowing operators to move away from static deployments and redefining fleet management.
You can read the full details of the methodology and findings here: Link: https://www.sciencedirect.com/science/article/pii/S2210539526002877