Where Ridehail Drivers Go Between Trips: Trading off Congestion and Curb Availability?

Millard-Ball, Adam; Liu, Liwei; Hansen, Whitney; Cooper, Drew; Castiglione, Joe · 2021 · ROSA P / University of California Institute of Transportation Studies

archive: archived pipeline: cataloged verified

Get this paper ↗ (full text — opens at the source; we link to it, we don't host it)

Summary

This study investigates the behavior of ridehail drivers (Uber and Lyft) when they are out of service between paid trips, addressing a gap in transportation research that has historically focused on passenger trips rather than driver deadheading. The authors aim to quantify how drivers trade off congestion and curb availability by choosing between cruising, repositioning, and parking. Understanding these behaviors is critical for assessing the environmental impacts of ridehailing, including vehicle miles traveled, congestion, and pollution, as well as for informing municipal policies regarding curbspace allocation and pricing. The researchers analyzed a dataset of 5.3 million ridehail trips in San Francisco collected between November and December 2016 via Uber and Lyft APIs. They developed a classification method to partition out-of-service segments into three categories: parking (vehicles stationary for at least three minutes), cruising (circling or backtracking), and repositioning (moving to a new location in anticipation of demand). Using map-matched GPS data and traffic speed information to distinguish parking from congestion, they applied multinomial logistic regression to identify factors influencing driver choices, incorporating variables such as household density, employment density, parking capacity, time of day, and neighborhood demographics. The results indicate that out-of-service travel accounts for 19% of total ridehail vehicle travel, with an average duration of 4.1 minutes and distance of 1.0 km per segment. Repositioning dominates driver behavior, accounting for 63% of out-of-service time, followed by cruising (23%) and parking (14%). Regression analysis reveals that drivers make rational, demand-responsive choices: they reposition away from dense residential areas during peak afternoon and evening hours when demand is higher elsewhere, but remain in those neighborhoods during mornings and nights. However, the study also finds suggestive evidence of racial bias, with drivers tending to avoid neighborhoods with high proportions of residents of color. Additionally, more experienced drivers cruise less, indicating that drivers learn over time that cruising is a suboptimal strategy. The findings have significant implications for urban policy. While repositioning reduces pressure on curbspace and improves fleet efficiency, it increases vehicle travel and associated externalities. Conversely, parking reduces vehicle miles but competes for limited curb space. The authors argue that current per-trip fees are insufficient to address these externalities. They recommend that cities implement distance- and time-based fees for all ridehail vehicle travel, regardless of passenger status, to internalize the costs of congestion and pollution. Furthermore, they suggest that app-based interventions could reduce cruising, and that place-based time charges could encourage drivers to park in areas where they do not conflict with other curbspace users.

Key finding

Repositioning accounts for 63 percent of out-of-service time, while drivers exhibit rational location choices based on demand but also show evidence of avoiding neighborhoods with high proportions of residents of color.

Methodology

dataset

Sample size: 5300000

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed. Discovered via bulk_ingest_rosap on 2026-05-23 (6 acquisition events logged).

Summary generated by qwen3.6-27b-prismaquant on 2026-06-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• exposure measurement