University of Limerick Institutional Repository

A synthesis of predictive statistics and actuarial science to quantify emerging and existing risks within the motor vehicle industry

DSpace Repository

Show simple item record

dc.contributor.advisor Murphy, Finbarr
dc.contributor.advisor Rizzi, Luiz
dc.contributor.advisor Mullins, Martin Shannon, Darren 2021-04-06T10:46:11Z 2021-04-06T10:46:11Z 2020
dc.description peer-reviewed en_US
dc.description.abstract The broad evolutionary movement toward ensuring vehicle safety has necessitated the collection motor vehicle collision (MVC) data, so that the circumstances that influence the frequency and severity of MVCs can be understood. The data generated from MVCs has resulted in the development of practical safety mechanisms within vehicles. Technological advancements such as telematics devices have the ability to transform driving behaviour, while advanced driver assistance systems (ADASs) have the ability to avoid or mitigate the severity of MVCs. The increased dissemination of vehicles equipped with these technologies will shift the dynamics of risks faced by road users. Consequently, claim and compensation patterns will be disrupted by these advancements as they transform the typology and causes of MVCs. As such, we propose in this thesis a number of proactive solutions that can be found using this influx of MVC data. The future of the motor insurance industry hinges on the efficient use of the magnitude of data that will become available with these technologies, so that accurate assessments of risk can be made for insured vehicles. This thesis contributes a number of methodological approaches that investigate the risks faced by road users and insurance providers alike. We further assess the role of primary insurers in a data-laden world. Chapters 2 and 3 review the methodological approaches that have traditionally been used to capture the severity of motor vehicle collisions (MVCs), and propose alternative approaches that allow the economic costs of MVCs to be directly related with the collision. These chapters primarily focus on the link between injury severity and economic cost, and use this information to discern the collision factors that influence the economic costs that are typically paid out in compensation claims. The results link aspects of insurance loss-event literature, injury severity literature, and MVC analysis literature in order to mitigate the impact of litigation risk faced by primary insurers. The results also point road safety practitioners to a number of collision factors that incur significant economic detriment. Chapter 4 explores the impact of relative impact velocity (delta-V) in an MVC, the collision factor that Chapter 2 and 3 identify as most influencing injury severity. A novel statistical approach is used to examine the intervening role that delta-V has between collision factors and MVC severity, through the lens of two injury severity metrics. The results highlight that a number of collision factors only influence injury severity because of the underlying role of relative impact velocity. The models generated also perform well in out-of-sample testing. Chapter 5 presents an occupant-focused approach to determine the collision events that are primarily linked with whiplash-related injuries – a pressing issue in the Irish and UK insurance arena due to the high frequency of compensation claims that are centred on minor cervical strains. We propose in this chapter a robust methodology that assigns whiplash likelihood estimates to drivers that are injured in MVCs, and compare the results of this methodology with realised incidents. Finally, Chapter 6 reflects on the current state of the motor insurance market and details the expected changes in actuarial considerations as ADAS-enabled vehicles, semi-autonomous vehicles (SAVs) and eventually, fully autonomous vehicles (AVs), become a common feature in the transport environment. Based on a multitude of factors that will present as advanced-technology vehicles become increasingly proliferated, it becomes clear that the actuarial impact that AV technology will have may not align with the actuarial considerations upon which the insurance industry currently operates. The discussions we provide in this chapter are beneficial as they spark a discussion on the future of actuarial science, and detail the inevitable shift toward reinsurers as key stakeholders of the motor insurance industry. The quantitative and qualitative assessments of risk provided in this thesis contribute to the field of transportation safety and insurance mathematics as they explore the risks faced by primary insurers and road users alike. The chapters within this thesis offer proactive solutions that can be used by primary insurers to mitigate the impact of these risks. Furthermore, the chapters offer insights in to the dynamics of collision events that influence injury severity, which may better inform road safety policies and vehicle engineering. en_US
dc.language.iso eng en_US
dc.publisher University of Limerick en_US
dc.subject motor vehicle industry en_US
dc.title A synthesis of predictive statistics and actuarial science to quantify emerging and existing risks within the motor vehicle industry en_US
dc.type info:eu-repo/semantics/doctoralThesis en_US
dc.type.supercollection all_ul_research en_US
dc.type.supercollection ul_theses_dissertations en_US
dc.rights.accessrights info:eu-repo/semantics/openAccess en_US

Files in this item

This item appears in the following Collection(s)

Show simple item record

Search ULIR


My Account