A clear understanding of the crash risks associated with care sharing might allow insurance companies to tailor their products to drivers and encourage more car sharing without compromising safety.
A team from Lithuania writing in the International Journal of Risk Assessment and Management has looked at the growth of car sharing in recent years in the context of insurance and risk. Obviously, there are significant benefits to car sharing in that it can reduce the total number of vehicles on the roads, reduce pollution and fuel use, and perhaps even reduce the number of road traffic accidents. Kristina Sutiene and Monika Uselyte of the Department of Mathematical Modeling at Kaunas University of Technology in Kaunas, Lithuania, have retrieved data from car-sharing systems and used linear regression and machine learning methods, such as regression trees and random forests, to model crash risk based on those observations.
They found that the average daily trip duration, the month of any crash event, and the make of car correlate most closely with the incidence of vehicle accidents. Holiday periods, working day or weekend, and peak hour had no valuable information for predicting crash risk. Additional the driver’s gender had no bearing on crash risk either.
The conventional approach taken by insurance companies in setting premiums usually takes into account the age, gender and other demographic factors of the driver to be insured as well as the make, value, and power of the vehicle. In the sharing economy, this model is somewhat outdated where experienced drivers of any gender may be equally as safe or otherwise. Moreover, there are many other factors that might be better predictors of crash risk as the Kaunas University of Technology team suggests, especially in the context of car sharing.
The team suggests that “After a proper assessment of the risk indicators that have the greatest impact on the occurrence of crashes, companies might be able to enter into personalised car-sharing pricing by developing usage-based or pay-as-you-drive insurance products.”
Sutiene, K. and Uselyte, M. (2021) ‘Factors affecting crash risk within the car-sharing market’, Int. J. Risk Assessment and Management, Vol. 24, Nos. 2/3/4, pp.236–251.