Technology is Reshaping Vehicle Insurance
Usage-based insurance (UBI) is a largely untapped market that is ripe for technology solutions to deliver tangible benefits to consumers, the insurance industry and to society.
The market for usage-based insurance (UBI) is largely untapped because the way in which automobile insurance risks were assessed stood still for many years. It was based on static, statistical data like age, gender, car model and so on. Applying telematics technology to usage-based insurance is a relatively recent development and it is delivering the means for insurers to make objective assessments of risk profiles based on real-time, dynamic data like mileage, areas travelled, time of day, keeping to speed limits, engine RPM and fuel level as well as driver behavior. This driver-specific information can also be paired with publicly available data to identify road types and weather conditions.
There are two key benefits arising from this shift: for insurers, there’s the ability to detect and retain the majority of the lowest risk drivers. For drivers, particularly young drivers, it’s a way of getting significant discounts on their premiums. In addition, these assessment mechanisms work like psychological conditioning, where good driving behavior is rewarded and, in a way, bad behavior is punished—with higher premiums or loss of coverage. And since careful drivers have fewer accidents, this becomes a big benefit for society.
Technology Revved Up and Ready
Automotive SatNav systems employ similar real-time dynamic data and when these applications became available for download and use on smartphones they exploded in popularity. Therefore, creating UBI apps for smartphones was a logical development. Smartphones have the requisite functionality, which includes sensors to detect acceleration, braking and cornering, but insurers have usability and reliability issues similar to those from texting-and-driving.
In the U.S., insurers have concerns about the use of smartphone apps and this is reflected in the fact that solutions based on in-vehicle, on-board diagnostics (OBD) dongles have become the preferred option. These devices plug into the vehicle’s OBD-II service port, similar to the one shown in Figure 1a. In the U.S., cars have been equipped with these ports since 1996. (The main reasons for their preference are covered in the following section on “Dedicated in-vehicle devices.”) Danlaw’s device, which is known as a DataLogger, is shown in Figure 1b. This designation indicates that UBI data is logged (stored) in the device before being transmitted over a cellular network to the insurer.
Nate Bryer, VP of innovation at automotive engineering firm Danlaw says: “Smartphones still lack the ability to provide clean, consistent and accurate driving data that can be used in usage-based insurance programs. As of today, the only way to generate and gather the data needed for a UBI program is an embedded or self-installed device that is tied into a vehicle’s electronic system.” However, the ability to download a UBI app and run a free trial with an insurance company is filling the awareness gap that’s allowing insurers to establish a business relationship with young drivers that can be carried forward when they marry, take out mortgages or need additional insurance policies.
For insurers, smartphone apps are tools to collect data that they can use for risk assessment. Free UBI trials allow smartphone services to be employed as a “teaser” that: (a) introduces the concept; (b) allows drivers to see their driving behavior at the end of the trial; and (c) informs them about the potential reduction in their premium if they drive carefully.
|Figure 1a (left): Typical OBD-II Port is easily accessible under the car dashboard. Figure 1b (right): Danlaw’s DataLogger 7-Series is a small, self-installed, OBD connected telematics solution.|
Seen in this context, smartphones are providing a valuable service, but their use in the UBI space is problematic for various reasons. For example, the device may not always be on every time the car is driven or the app may not be compatible and certified for use with the phone’s operating system or platforms, both of which vary dramatically across the consumer industry. There are vocal advocates for smartphone UBI, but the insurance industry and regulators are key players and particularly concerned about operational reliability and fraud. In addition, offering a UBI policy that requires ownership of a smartphone is unlikely to pass the “fairness” test, as many drivers do not own one of these pricey devices.
As well, regulators are highly critical of the reliability of data delivery. Phones can be removed accidentally or run out of battery power. Users would need to start the UBI app manually and phones could be dropped or become airborne during an impact. Telematics behavioral data such as braking, turning and accelerating is likely to be inaccurate and unreliable because phones are rarely perfectly oriented—data must originate from sensors oriented with the vehicle’s travel plane. There are a number of other issues, too, with the most significant being liability exposure.
In almost all international jurisdictions, courts will find companies liable for negligence and damages when there is a “better” solution than the smartphone (e.g., employing a data logger) that the company should have considered. This reliability concern is more critical in life-saving applications such as an emergency response request in case of a crash, but any automotive application is held liable for delivering what it promises, even if the promise is lower insurance rates. Moreover, the smartphone UBI scenario could adversely impact on the validity of fair and honest claims. Only an embedded device can guarantee accurate X=Y orientation of high g-force data, which is needed to justify insurance claims for whiplash.
When designing a smartphone policy, insurers will need to look at three unknowns. The first is the service itself and the reason why the driver will want to use the app. The second unknown concerns the drivers, since it will never be possible to control how and if they use the app. The third is the device itself, since each smartphone is equipped with different sensor grades and quality. There is nothing to legislate the driver’s use of the app. The UBI app will gather second-by-second information and build a picture of the driver’s behavior, but it will not match what a data logger can do. The insurer has to accept that risk and include it in the model from the start.
Dedicated In-Vehicle Devices
In the U.S., insurers’ concerns about smartphones are reflected in the fact that solutions based on in-vehicle OBD-II data loggers have become the preferred option. These robust devices are unobtrusive and because they have a semi-permanent wired interface to the vehicle’s electronic system, they provide accurate driving data. They are relatively cheap, there is no installation cost, nor is there any need to schedule an installation appointment. They can also be used in tandem with a smartphone. This hybrid solution would combine the data quality of the installed device with customer-friendly features like on-screen displays, which could include warning flags about bad driving and the possible financial impact.
|Dedicated in-vehicle devices that employ high-frequency sampling provide accurate observations of a driver’s behaviour. They can identify bad behaviour such as tailgating (Figure 2a) – see how brown line shows excessive speed changes; and detect details like gear changes (Figure 2b bottom). This type of detailed information is particularly useful when accidents need to be reconstructed. Purple shows data from GPS data at 1 sample per second and Brown is high-frequency sampling of speed sensor data – Graphics courtesy Redtail Telematics.|
UBI solutions based on OBD-II in-vehicle devices address the concerns of the insurance industry and regulators. For example:
Fairness. Regardless of vehicle type, demographics or socio-economic status, all insured drivers are measured the same way.
Reliability. A dedicated hardware solution ensures that the connectivity between the vehicle and provider is controllable and timely. And an embedded in-car system can distinguish a real crash and emergency from the bumps of normal driving (potholes, climbing curbs, speed bumps and road transitions).
Security. Dedicated hardware eliminates the potential for fraud, as it is typically a proprietary system that relies on a direct connection to the insured’s vehicle.
No-distraction driving. Governments around the world are grappling with the potentially disastrous consequences of cell-phone-induced distraction while driving. In 2010, this type of distraction was directly linked to over 3,000 fatal car accidents in the U.S. Insurers cringe at the thought of being dragged into court if law enforcement finds that drivers are causing accidents while interacting with their UBI smartphone apps.
Storing and Analyzing the Data
UBI can quickly accumulate massive amounts of data at the petabyte level and beyond. If a vehicle were driven 1,000 miles a month, it would typically generate over 190K data points a year. Insuring 1,000 drivers, which is a modest figure, would take this figure to over 190 million data points. You can do the math. However, data loggers can employ processing power-enabling smart filtering of the raw data. Only the relevant data is transmitted over-the-air and sorting can also be employed to ensure that the data plan stays low. Data loggers employing Telit modules already have this capability.
Nevertheless, a lot of UBI “Big Data” is going to be generated in the coming years and insurers need a way of scoring driving behavior and then applying scores to rating algorithms to reward drivers who drive safely. This is not a trivial task but it is one that the larger insurers can accommodate; others will typically partner with companies specialized in data management and predictive modelling.
Storing and analyzing Big Data is a generic issue for other M2M sectors and other industries and we are witnessing a rise in the number of innovative, cloud-based solutions. They include visual analytics, which would allow driving behavior of all insured drivers to be presented in an easy-to-understand, graphical interface.
It is clear that smartphone popularity and penetration will continue to rise in in the coming years with a mind-boggling number of apps that can be downloaded. However, the use of UBI smartphone apps is problematic. The regulatory climate is unfavorable and the legal risks are significant. Dedicated solutions based on proprietary, dedicated hardware that are embedded or connected to the vehicle’s OBD-II diagnostic port provide robust results that can accommodate the demanding requirements of the insurance industry and the regulators. As illustrated in figure 2a and 2b, they provide very accurate driving data and the processing power of a dedicated hardware solution enables smart filtering of raw data. Also, they are as easy to install by the insured driver as downloading a UBI app into the smartphone.
Cyril Zeller is senior sales director, Global Telematics Segment at Telit Wireless Solutions. He is responsible for developing and executing Telit’s corporate strategy for the worldwide telematics industry, especially in the area of fleet management, stolen vehicle recovery and usage-based insurance. Prior to joining Telit, Zeller served as the vice president of sales and marketing for Mobile Devices Ingénierie, a European-based leader in open-platform telematics technology. Zeller is an expert in the financial and legal issues in telematics.