Exploring the determinants of bike-sharing member trips with explainable machine learning: insights from Chicago’s Divvy system

Bike-sharing offers an eco-friendly and efficient transportation alternative, addressing traffic congestion, environmental concerns, and promoting healthier lifestyles. Understanding the factors that influence bike-sharing trips made by members is essential for the long-term operational and financial sustainability of bike-sharing systems. However, existing studies have primarily relied on survey data and traditional statistical models, which are often constrained by limited sample sizes and an inability to fully capture non-linear relationships or complex interactions between various factors and bike-sharing membership trips. To address these shortcomings, this study leverages million-level trip records from Chicago’s Divvy bike-sharing system and applies explainable machine learning model to identify key determinants influencing bike-sharing trips made by members. Unlike traditional survey-based studies, the proposed framework exploits large-scale operational data to capture complex behavioral patterns and interactions that are not readily observable using standard models. The results indicate a strong positive association between membership-based bike-sharing trips and weekday morning and afternoon commuting periods. Additional factors, such as longer travel distances, proximity to downtown, higher proportions of younger populations, and greater shares of non-White residents, are also positively associated with membership trip activity. These findings both corroborate and extend existing literature, while providing new, data-driven insights enabled by advanced analytical methods. Based on the identified determinants, the study proposes targeted policy and operational strategies to encourage bike-sharing membership adoption. By explicitly overcoming prior methodological limitations, this research offers a robust and scalable analytical framework to support bike-sharing operators in service planning, membership expansion, and the sustainable growth of urban mobility systems.