Assessing Driving Risk Using Internet of Vehicles Data: An Analysis Based on Generalized Linear Models

Are electric vehicles riskier? A comparative study of driving behaviour and insurance claims for internal combustion engine, hybrid and electric vehicles

Electric vehicles (EVs) differ significantly from their internal combustion engine (ICE) counterparts, with reduced mechanical parts, Lithium-ion batteries and differences in pedal and transmission control. These differences in vehicle operation, coupled with the proliferation of EVs on our roads, warrant an in-depth investigation into the divergent risk profiles and driving behaviour of EVs, Hybrids (HYB) and ICEs. In this unique study, we analyze a novel telematics dataset of 14,642 vehicles in the Netherlands accompanied by accident claims data. We train a Logistic Regression model to predict the occurrence of driver at-fault claims, where an at-fault claim refers to First and Third Party damages where the driver was at fault. Our results reveal that EV drivers are more exposed to incurring at-fault claims than ICE drivers despite their lower average mileage. Additionally, we investigate the financial implications of these increased at-fault claims likelihoods and have found that EVs experience a 6.7% increase in significant first-party damage costs compared to ICE. When analyzing driver behaviour, we found that EVs and HYBs record fewer harsh acceleration, braking, cornering and speeding events than ICE. However, these reduced harsh events do not translate to reducing claims frequency for EVs. This research finds evidence of a higher frequency of accidents caused by Electric Vehicles. This burden should be considered explicitly by regulators, manufacturers, businesses and the general public when evaluating the cost of transitioning to alternative fuel vehicles.

Pricing weekly motor insurance drivers' with behavioral and contextual telematics data

Telematics boxes integrated into vehicles are instrumental in capturing driving data encompassing behavioral and contextual information, including speed, distance travelled by road type, and time of day. These data can be amalgamated with drivers' individual attributes and reported accident occurrences to their respective insurance providers. Our study analyzes a substantial sample size of 19,214 individual drivers over a span of 55 weeks, covering a cumulative distance of 181.4 million kilometers driven. Utilizing this dataset, we develop predictive models for weekly accident frequency. As anticipated based on prior research with yearly data, our findings affirm that behavioral traits, such as instances of excessive speed, and contextual data pertaining to road type and time of day significantly aid in ratemaking design. The predictive models enable the creation of driving scores and personalized warnings, presenting a potential to enhance traffic safety by alerting drivers to perilous conditions. Our discussion delves into the construction of multiplicative scores derived from Poisson regression, contrasting them with additive scores resulting from a linear probability model approach, which offer greater communicability. Furthermore, we demonstrate that the inclusion of lagged behavioral and contextual factors not only enhances prediction accuracy but also lays the foundation for a diverse range of usage-based insurance schemes for weekly payments.

Bibliometric review of telematics-based automobile insurance: Mapping the landscape of research and knowledge

Telematics technology and its implementation in auto insurance have received great interest due to their potential to transform the insurance sector and promote safer driving practices. By implementing telematics technology, insurers may tailor insurance premiums to individual drivers, taking into account their real driving habits and performance, ultimately leading to improved road safety, cost savings, and an empowered driving community. The current study, through bibliometric analysis, carefully identifies and evaluates the existing body of scholarly literature on this subject for the last 21 years, including journal articles, conference papers, and related publications. The analysis uncovers key research studies, influential authors, top publication outlets, top countries with collaborations, and prolific research fields, providing useful insights into the evolution and growth of telematics-based insurance research. Furthermore, thematic mapping, cluster analysis, and critical analysis of top recent studies aided in identifying key research clusters and themes, as well as potential gaps and areas for further exploration, guiding future researchers and policymakers in advancing this transformative technology.

The economic imperatives for technology enabled wellness centered healthcare

A 2020 World Health Organization report underscored the impact of rising healthcare spending globally and questioned the long-term economic sustainability of current funding models. Increases in costs associated with care of late-stage irreversible diseases and the increasing prevalence of debilitating neurodegenerative disorders, coupled with increases in life expectancy are likely to overload the healthcare systems in many nations within the next decade if not addressed. One option for sustainability of the healthcare system is a change in emphasis from illness to wellness centered care. An attractive model is the P4 (Predictive, Preventative, Personalized and Participatory) medicine approach. Recent advances in connected health technology can help accelerate this transition; they offer prediction, diagnosis, and monitoring of health-related parameters. We explain how to integrate such technologies with conventional approaches and guide public health policy toward wellness-based care models and strategies to relieve the escalating economic burdens of managed care.

Driving Risk Assessment Using Near-Miss Events Based on Panel Poisson Regression and Panel Negative Binomial Regression

This study proposes a method for identifying and evaluating driving risk as a first step towards calculating premiums in the newly emerging context of usage-based insurance. Telematics data gathered by the Internet of Vehicles (IoV) contain a large number of near-miss events which can be regarded as an alternative for modeling claims or accidents for estimating a driving risk score for a particular vehicle and its driver. Poisson regression and negative binomial regression are applied to a summary data set of 182 vehicles with one record per vehicle and to a panel data set of daily vehicle data containing four near-miss events, i.e., counts of excess speed, high speed brake, harsh acceleration or deceleration and additional driving behavior parameters that do not result in accidents. Negative binomial regression (AICoverspeed = 997.0, BICoverspeed = 1022.7) is seen to perform better than Poisson regression (AICoverspeed = 7051.8, BICoverspeed = 7074.3). Vehicles are separately classified to five driving risk levels with a driving risk score computed from individual effects of the corresponding panel model. This study provides a research basis for actuarial insurance premium calculations, even if no accident information is available, and enables a precise supervision of dangerous driving behaviors based on driving risk scores.

Boosting Poisson regression models with telematics car driving data

With the emergence of telematics car driving data, insurance companies have started to boost classical actuarial regression models for claim frequency prediction with telematics car driving information. In this paper, we propose two data-driven neural network approaches that process telematics car driving data to complement classical actuarial pricing with a driving behavior risk factor from telematics data. Our neural networks simultaneously accommodate feature engineering and regression modeling which allows us to integrate telematics car driving data in a one-step approach into the claim frequency regression models. We conclude from our numerical analysis that both classical actuarial risk factors and telematics car driving data are necessary to receive the best predictive models. This emphasizes that these two sources of information interact and complement each other.

Bivariate Mixed Poisson and Normal Generalised Linear Models with Sarmanov Dependence—An Application to Model Claim Frequency and Optimal Transformed Average Severity

The aim of this paper is to introduce dependence between the claim frequency and the average severity of a policyholder or of an insurance portfolio using a bivariate Sarmanov distribution, that allows to join variables of different types and with different distributions, thus being a good candidate for modeling the dependence between the two previously mentioned random variables. To model the claim frequency, a generalized linear model based on a mixed Poisson distribution -like for example, the Negative Binomial (NB), usually works. However, finding a distribution for the claim severity is not that easy. In practice, the Lognormal distribution fits well in many cases. Since the natural logarithm of a Lognormal variable is Normal distributed, this relation is generalised using the Box-Cox transformation to model the average claim severity. Therefore, we propose a bivariate Sarmanov model having as marginals a Negative Binomial and a Normal Generalized Linear Models (GLMs), also depending on the parameters of the Box-Cox transformation. We apply this model to the analysis of the frequency-severity bivariate distribution associated to a pay-as-you-drive motor insurance portfolio with explanatory telematic variables.

An Acceptance Approach for Novel Technologies in Car Insurance

Background: Unlike other financial services, technology-driven changes in the insurance industry have not been a vastly explored topic in scholarly literature. Incumbent insurance companies have hitherto been holding their positions using the complexity of the product, heavy regulation, and gigantic balance sheets as paramount factors for a relatively slow digitalization and technological transformation. However, new technologies such as car telematic devices have been creating a new insurance ecosystem. The aim of this study is to assess the telematics technology acceptance for insurance purposes. Methods: The study is based on the Unified Theory of Acceptance and Use of Technology (UTAUT). By interviewing 502 new car buyers, we tested the factors that affect the potential usage of telematic devices for insurance purposes. Results: The results indicate that facilitating conditions are the main predictor of telematics use. Moreover, privacy concerns related to the potential abuse of driving behavior data play an important role in technology acceptance. Conclusions: Although novel insurance technologies are mainly presented as user-driven, users (drivers and insurance buyers) are often neglected as an active party in the development of such technologies.

A Sarmanov Distribution with Beta Marginals: An Application to Motor Insurance Pricing

Background: The Beta distribution is useful for fitting variables that measure a probability or a relative frequency. Methods: We propose a Sarmanov distribution with Beta marginals specified as generalised linear models. We analyse its theoretical properties and its dependence limits. Results: We use a real motor insurance sample of drivers and analyse the percentage of kilometres driven above the posted speed limit and the percentage of kilometres driven at night, together with some additional covariates. We fit a Beta model for the marginals of the bivariate Sarmanov distribution. Conclusions: We find negative dependence in the high quantiles indicating that excess speed and night-time driving are not uniformly correlated.