A survey on driving behavior analysis in usage based insurance using big data
DOI: 10.1186/s40537-019-0249-5
archive: archived pipeline: cataloged verified
Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)
Summary
This survey paper addresses the evolution of Usage-Based Insurance (UBI) in the automotive sector, specifically focusing on the integration of big data analytics to assess driving behavior. The authors identify a critical gap in existing UBI models: while early Pay-As-You-Drive (PAYD) models relied solely on mileage and Pay-How-You-Drive (PHYD) models assessed post-trip metrics like speeding and hard braking, neither adequately addressed real-time risk mitigation or emotional factors such as aggressive driving and road rage. The research is motivated by the rising global incidence of road accidents, largely attributed to human behavioral factors, and the exponential growth of telematics data that necessitates advanced processing capabilities. The study employs a comprehensive literature review and analysis of existing industry solutions and academic research to categorize the third generation of UBI: Manage-How-You-Drive (MHYD). The authors examine data collection methodologies, including black boxes, dongles, embedded telematics, and smartphone-based sensors (GPS and inertial sensors). They classify MHYD into four distinct monitoring categories: driving pattern monitoring, fatigue monitoring, drowsiness detection, and driver distraction detection. The paper evaluates various machine learning algorithms and data parameters used in prior studies, such as Support Vector Machines, Neural Networks, and Hidden Markov Models, noting limitations in previous works regarding data volume, reliance on simulated data, or lack of real-time alerting mechanisms. Key findings highlight that while PHYD models have stabilized, MHYD remains an evolving field requiring more robust solutions for proactive engagement. The survey reveals that existing industry implementations, such as those by Allstate, Progressive, and TD Insurance, often focus on post-trip analysis or limited real-time alerts (e.g., speed only). The authors argue that current systems fail to sufficiently detect aggressive driving and road rage, which are significant precursors to accidents. Consequently, the paper proposes a novel solution framework that utilizes big data technologies to process high-velocity GPS and sensor data in real-time. This proposed model aims to detect aberrations by considering both behavioral metrics (acceleration, braking, cornering) and emotional factors, providing immediate alerts to drivers to prevent risky incidents. The significance of this work lies in its contribution to the development of more accurate, personalized insurance premiums and enhanced driver safety. By shifting from reactive post-trip analysis to proactive real-time management, the proposed MHYD approach bridges the gap between insurers and customers. The authors conclude that integrating big data analytics with machine learning to monitor aggressive and emotional driving behaviors will allow insurers to assess risk more precisely, reduce claim frequencies, and promote safer driving habits through immediate feedback, thereby advancing the maturity of the UBI ecosystem.
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 | OpenAlex-citations | — | — | 1 | 2026-06-25 |
| archive | success | unpaywall | — | — | 2 | 2026-06-26 |
| extract | success | cached | — | — | 2 | 2026-06-26 |
| clean | success | clean | — | — | 1 | 2026-06-25 |
| chunk | success | chunk | — | — | 1 | 2026-06-25 |
| embed | success | embed | Qwen/Qwen3-Embedding-8B | — | 1 | 2026-06-25 |
| promote | success | — | — | — | 1 | 2026-06-25 |
| summarize | success | llm | qwen3.6-27b-prismaquant | summ-v5 | 1 | 2026-06-26 |
| tag | success | vector_similarity | — | — | 6 | 2026-06-25 |
| verify | partial | — | — | — | 1 | 2026-06-26 |
Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified_with_issues.
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: observational prevalence
- Methodological Resource: dataset resource
- Theoretical Contribution: computational model