Forensic Examinations in Fraud Cases in the Field of Auto Insurance

The proposed article considers the possibilities of various types of expert research on criminal cases of fraud in the field of automobile insurance (CTP, CASCO). On the example of illustrative cases from the published judicial practice in criminal cases, the features of the appointment of certain types of forensic examinations, their research potential are demonstrated. Clearly shows the wide the possibility of applying special knowledge to improve the effectiveness of detecting and investigating insurance fraud committed by staging and falsifying the circumstances of road accidents. The prospects for the use of atypical forensic examinations in criminal cases of relevant crimes are shown. The article is recommended not only to employees of investigative departments of law enforcement agencies, judges and experts, but also to a wide range of readers interested in countering fraud and other crimes committed in the insurance industry.

A Blockchain-Based Auto Insurance Data Sharing Scheme

Auto electronic insurance policy and electronic maintenance list record the entire process of auto owners purchasing auto insurance and repairs after accident, respectively. They play a vital role in auto owners’ applications for claims and insurance company’s judgment on whether to settle the claims. However, the privacy of insurance policy and the “information island” resulting from the nonsharing of data between users make the claim has low efficiency. The notable features of blockchain technology are decentralization and tamper-proof, which can well solve data sharing and privacy protection. This paper proposes a blockchain-based auto insurance data sharing scheme to improve the existing auto insurance claim system. The scheme includes four main bodies: auto owner, insurance company, 4S Shop, and government authority. In the proposed scheme, the data sharing of authorized users is realized through proxy reencryption. Finally, we have analyzed the security and performance of the solution. The analysis results show that the proposed scheme can meet many security features such as user access control and data tamper resistance and has an ideal calculation and communication cost.

Prediction of neck injury using multiple linear regression and bumper damage: Case study for low-speed rear-end collisions

The most frequent type of traffic accident is a low-speed rear-end collision, which can damage parts of the vehicle, including the bumper, and cause neck injury to the occupants. Even in minor damage accidents, such as scratches on bumper covers, 26.3% of occupants received treatment for bodily injuries whose main symptom was neck injuries through auto insurance. This study was conducted to evaluate the potential for neck injuries in low-speed accidents. Fifty-nine low-speed rear-end impact tests were conducted, and the motion of the struck vehicle and the neck injury criterion (NIC) of the occupant according to the test conditions were predicted using multiple linear regression derived via supervised machine learning. It was confirmed that the NIC can be predicted using vehicle motion values that can be obtained through an event data recorder. The coefficients of determination of the regression equations were 0.67–0.83. Lastly, we investigated whether neck injuries can be predicted through bumper cover damage that can be checked immediately after a vehicle accident. In the case of the vehicle damage type 1/2/3 category applied to auto insurance by the Korean government, an occupant would have a very low possibility of neck injury or symptoms. No symptoms or injuries were reported in the volunteer tests conducted for this study.

Improving Explainability of Major Risk Factors in Artificial Neural Networks for Auto Insurance Rate Regulation

In insurance rate-making, the use of statistical machine learning techniques such as artificial neural networks (ANN) is an emerging approach, and many insurance companies have been using them for pricing. However, due to the complexity of model specification and its implementation, model explainability may be essential to meet insurance pricing transparency for rate regulation purposes. This requirement may imply the need for estimating or evaluating the variable importance when complicated models are used. Furthermore, from both rate-making and rate-regulation perspectives, it is critical to investigate the impact of major risk factors on the response variables, such as claim frequency or claim severity. In this work, we consider the modelling problems of how claim counts, claim amounts and average loss per claim are related to major risk factors. ANN models are applied to meet this goal, and variable importance is measured to improve the model’s explainability due to the models’ complex nature. The results obtained from different variable importance measurements are compared, and dominant risk factors are identified. The contribution of this work is in making advanced mathematical models possible for applications in auto insurance rate regulation. This study focuses on analyzing major risks only, but the proposed method can be applied to more general insurance pricing problems when additional risk factors are being considered. In addition, the proposed methodology is useful for other business applications where statistical machine learning techniques are used.