A “pay-how-you-drive” car insurance approach through cluster analysis

Carfora, Maria Francesca; Martinelli, Fabio; Mercaldo, Francesco; Nardone, Vittoria; Orlando, Albina; Santone, Antonella; Vaglini, Gigliola

doi:10.1007/s00500-018-3274-y

Cited by 56 publications

(35 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The relationship between accident risk has been found in previous studies. So, the level of braking pulse intensity can be related positively with proportionally higher insurance prices (Bian et al, 2018;Carfora et al, 2019;Tselentis, 2016 and2017).…”

Section: Discussionmentioning

confidence: 99%

Regression scores to identify risky drivers from braking pulses

Sun

Guillén

et al. 2020

CARMA 2020 - 3rd International Conference on Advanced Research Methods and Analytics

View full text Add to dashboard Cite

Driving data record information on style and patterns of vehicles that are inmotion. These data are analysed to obtain risk scores that can later beimplemented in insurance pricing schemes. Scores may also be used in on-board sensors to create risk alerts that help drivers to keep up with safetymargins. Regression methods are proposed and a prototype real sample of 253drivers is analysed. Conclusions are drawn on the mean number of brakepulses per day as measured within 30 seconds time-intervals. Linear andlogistic regressions serve to construct a label that classifies drivers. A novelfactor based on the driving range that is defined from geo-localizationimproves the results considerably. Driving range is expressed as measures thediagonal of a rectangle that contains the furthest North-South versus East-West weekly vehicle trajectory. This factor shows that frequent braking activityis negatively related to the square of driving range.

show abstract

Section: Discussionmentioning

confidence: 99%

Regression scores to identify risky drivers from braking pulses

Sun

Guillén

et al. 2020

CARMA 2020 - 3rd International Conference on Advanced Research Methods and Analytics

View full text Add to dashboard Cite

show abstract

“…In a connected car environment, the driver can access his/her car, which is further connected to the edge server that provides all driving-related services. These services include insurance services [ 1 , 2 ] by awarding scores to the drivers, providing traffic information, providing optimal route information [ 3 ] (based on fuel efficiency and shortest distance), manufacturer services [ 4 ] (e.g., predictive vehicle maintenance), and monitoring services. The increasing number of in-vehicle sensors and their connectivity with the driver and other vehicles via edge and cloud for multiple services is increasing the comfort level of society.…”

Section: Introductionmentioning

confidence: 99%

Lightweight Driver Behavior Identification Model with Sparse Learning on In-Vehicle CAN-BUS Sensor Data

Ullah

Kim

2020

Sensors

View full text Add to dashboard Cite

This study focuses on driver-behavior identification and its application to finding embedded solutions in a connected car environment. We present a lightweight, end-to-end deep-learning framework for performing driver-behavior identification using in-vehicle controller area network (CAN-BUS) sensor data. The proposed method outperforms the state-of-the-art driver-behavior profiling models. Particularly, it exhibits significantly reduced computations (i.e., reduced numbers both of floating-point operations and parameters), more efficient memory usage (compact model size), and less inference time. The proposed architecture features depth-wise convolution, along with augmented recurrent neural networks (long short-term memory or gated recurrent unit), for time-series classification. The minimum time-step length (window size) required in the proposed method is significantly lower than that required by recent algorithms. We compared our results with compressed versions of existing models by applying efficient channel pruning on several layers of current models. Furthermore, our network can adapt to new classes using sparse-learning techniques, that is, by freezing relatively strong nodes at the fully connected layer for the existing classes and improving the weaker nodes by retraining them using data regarding the new classes. We successfully deploy the proposed method in a container environment using NVIDIA Docker in an embedded system (Xavier, TX2, and Nano) and comprehensively evaluate it with regard to numerous performance metrics.

show abstract

“…Driving distance is one factor that has been widely explored (Lemaire et al, 2016;Boucher et al, 2017;Verbelen et al, 2018), other methods aim at evaluating driving risk based on extracting behavior variables from usage-basedinsurance (UBI) data that goes beyond driving distance (Bian et al, 2018;Ayuso et al, 2016a,b;Denuit et al, 2019). Carfora et al (2019) propose an indicator of driver aggressiveness based on cluster analysis results. More recently, generalised linear models are built based on the internet of vehicles (IoV) data to identify risky drivers, see Sun et al (2020).…”

Section: Introductionmentioning

confidence: 99%

Clustering driving styles via image processing

Zhu

Wüthrich²

2020

Ann. actuar. sci.

View full text Add to dashboard Cite

It has become of key interest in the insurance industry to understand and extract information from telematics car driving data. Telematics car driving data of individual car drivers can be summarised in so-called speed–acceleration heatmaps. The aim of this study is to cluster such speed–acceleration heatmaps to different categories by analysing similarities and differences in these heatmaps. Making use of local smoothness properties, we propose to process these heatmaps as RGB images. Clustering can then be achieved by involving supervised information via a transfer learning approach using the pre-trained AlexNet to extract discriminative features. The K-means algorithm is then applied on these extracted discriminative features for clustering. The experiment results in an improvement of heatmap clustering compared to classical approaches.

show abstract

A “pay-how-you-drive” car insurance approach through cluster analysis

Cited by 56 publications

References 27 publications

Regression scores to identify risky drivers from braking pulses

Regression scores to identify risky drivers from braking pulses

Lightweight Driver Behavior Identification Model with Sparse Learning on In-Vehicle CAN-BUS Sensor Data

Clustering driving styles via image processing

Contact Info

Product

Resources

About