Naturalistic Spatial Road Safety Analysis: The SmartMaps Project
DOI: 10.1007/978-3-031-88974-5_37
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 need for proactive road safety analysis by leveraging large-scale spatio-temporal data from smartphone sensors to evaluate Surrogate Safety Measures (SSMs). Traditional crash data often suffers from under-reporting and bias, whereas SSMs, such as harsh braking, offer automated, real-time indicators of safety risks before crashes occur. The research specifically aims to analyze the spatial distribution of harsh braking events across various road environments in the Western Greece Region, utilizing geometric design characteristics and naturalistic driving behavior data. The methodology involved analyzing 9,355 road segments with an average length of 223 meters, derived from OpenStreetMap data. The dataset included 14,161 trips recorded in 2021 via the OSeven Telematics smartphone application. The study compared two statistical models: a non-spatial log-linear model and a spatial error model, both using the logarithm of harsh braking events as the dependent variable. Independent variables included trip count, segment length, linearity index, speeding duration, mobile phone usage duration, and road type. The spatial error model was selected to account for spatial autocorrelation in residuals, allowing for a more robust assessment of spatial dependencies. The results demonstrated that both models yielded consistent coefficient signs, indicating positive correlations between harsh braking events and risk exposure indicators such as segment length and trip count. Additionally, speeding, mobile phone usage, and higher road segment linearity (fewer curves) were positively associated with increased harsh braking. Conversely, motorways exhibited significantly lower harsh braking rates compared to other road types. Crucially, the spatial error model outperformed the non-spatial log-linear model, evidenced by a lower Akaike Information Criterion (AIC) value of 11,824 versus 11,826 and a statistically significant spatial parameter (Lambda = 0.016, p=0.041). This confirms that accounting for spatial autocorrelation improves model fit and reliability. The significance of this research lies in validating harsh braking as a reliable SSM for proactive safety assessment and demonstrating the superiority of spatial models in analyzing telematics data. By identifying specific behavioral and geometric factors contributing to safety risks, the findings support the development of comprehensive mapping tools, such as the SmartMaps project, aimed at promoting safe and eco-friendly driving behaviors. The study underscores the value of integrating smartphone-derived naturalistic data with spatial statistical techniques to enhance road safety management and countermeasure evaluation.
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 | Crossref | — | — | 1 | 2026-06-18 |
| archive | success | canonical_url | — | — | 1 | 2026-06-25 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-18 |
| chunk | success | chunk | — | — | 1 | 2026-06-18 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-18 |
| promote | success | — | — | — | 1 | 2026-06-18 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-18 |
| verify | success | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.
Topics
Ranked by relevance to this paper. Hover a topic for its definition.
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