Driving safety assessment for ride-hailing drivers
DOI: 10.1016/j.aap.2020.105574
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This study addresses the safety risks associated with ride-hailing drivers, a population characterized by significant heterogeneity in experience and operational habits compared to traditional taxi drivers. Motivated by the rapid growth of ride-hailing services and limited research on driver-specific crash risk factors, the authors aimed to identify and quantify the impact of driving history and operational characteristics on crash rates. The research seeks to provide data-driven insights for developing safety countermeasures, driver education programs, and regulatory policies within the industry. The analysis utilized a cross-sectional dataset of 189,815 active drivers from Didi Chuxing Technology Corporation in a major Chinese city during the second half of 2018. These drivers completed over two billion vehicle-kilometers and were involved in 5,298 crashes. The study extracted seven operational features, including crash history, total booking distance, years of service, average passenger rating, percentage of long-shift bookings, and operations during morning or evening peak hours. To account for potential nonlinear relationships between risk factors and crash rates, the authors employed a Poisson Generalized Additive Model (GAM). Additionally, the SHapley Additive exPlanation (SHAP) method was used to assess and visualize the contribution of each factor to individual and population-level crash risk. The results identified several factors significantly associated with crash risk. Crash history, the percentage of long-shift bookings, total driving distance, operations during peak hours, years of being a ride-hailing driver, and passenger ratings all showed significant associations. Notably, the relationship between the percentage of long-shifts and crash risk was nonlinear; crash rates decreased for drivers with moderate long-shift percentages but increased sharply for those with high percentages, reflecting a mix of inexperienced, cautious, and professional drivers. Using SHAP values to rank global impact, the study found that passenger average rating, total driving distance, and crash history were the leading contributing factors to crash risk. Drivers with prior crash involvement or lower passenger ratings exhibited substantially higher crash rates. The findings highlight the importance of considering nonlinear relationships and operational metrics in assessing ride-hailing driver safety. By identifying passenger ratings and crash history as primary risk indicators, the study suggests that ride-hailing platforms can leverage these readily available data points to target safety interventions. The results support the development of tailored safety education and regulatory frameworks that address specific risk profiles, such as drivers engaging in excessive long shifts or those with poor service ratings, thereby improving overall safety in the ride-hailing sector.
Key finding
Passenger average rating, total driving distance, and crash history were identified as the leading contributing factors to crash risk among ride-hailing drivers.
Methodology
dataset
Sample size: 189815
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.
| Stage | Outcome | Tool | Model | Prompt | Attempts | Completed |
|---|---|---|---|---|---|---|
| discover | success | — | — | — | 1 | 2026-05-28 |
| archive | success | canonical_url | — | — | 4 | 2026-06-06 |
| extract | success | cached | — | — | 3 | 2026-06-10 |
| clean | success | clean | — | — | 1 | 2026-06-07 |
| chunk | success | chunk | — | — | 1 | 2026-06-07 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-07 |
| enrich | success | semantic_scholar | — | — | 4 | 2026-07-02 |
| promote | success | — | — | — | 1 | 2026-06-04 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 2 | 2026-06-10 |
| tag | success | vector_similarity | — | — | 15 | 2026-06-11 |
| verify | success | — | — | — | 2 | 2026-06-10 |
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.
- telematics crash prediction
- exposure measurement
- induced exposure
- sex gender
- novice drivers
- incidence prevalence
Information type
What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).
- Empirical Findings: crash risk outcomes
- Methodological Resource: dataset resource
- Theoretical Contribution: computational model