One of the fastest growing spaces that take advantage of big data is telematics insurance, and it’s only expected to continue to grow as more “connected cars” come onto the market.
Telematics has actually been in use for more than 15 years, but it’s only with the burgeoning market of connected cars that it has actually taken off as an industry. Indeed, the global vehicle telematics industry is predicted to reach more than $100 billion dollars by 2022. It’s already estimated in the U.K. that nearly half a million drivers are choosing to have their driving monitored, and by 2020 there will be 220 million connected cars on the road, representing $2.3 trillion dollars in sales, according to data collected by Business Insider. Dell even bet $1B on IoT at the edge earlier this month.
Each connected vehicle is expected to produce an eye-popping 25 GB of data per hour, according to Hitachi Data Systems. Multiply that out by 220 million, and we’re looking at an explosion of data to the level few industries are currently equipped to manage.
What is Telematics Insurance?
Vehicle telematics, which involves monitoring the movement, distance and other metrics of a vehicle, are playing a key role in how insurance companies charge their customers.
Telematics insurance is car insurance where a telematics box is fitted to your car. It hen measures various aspects of how, when and where you drive. The data from the telematics box collects the time of day or night you drive, the speed you drive at on different sorts of roads, if you brake or accelerate sharply if you take breaks on long journeys, your motorway miles, and your total mileage.
The information collected is then used to assess your car insurance risk, help calculate the cost of your renewal premium, give safe drivers Bonus Miles each month, manage your claim after an accident and more.
Real-time Examples of Connected Cars
If you’re in the U.S., for example, you may have seen TV ads for Progressive Insurance’s Snapshot program, which uses a mobile app or plug-in device to record driver behavior and then translate that into personalized insurance rates. According to a Quartz article that was published when the program was still in its infancy, Progressive found that driving behaviors such as “actual miles driven, braking and time of day driving” are better predictors of an accident than traditional insurance rating variables such as age, gender, and vehicle.
Another example is the Volvo connected cars. They started off with telemetry systems in Sweden, which meant that if the car detected ice on the road because they put the electronic braking distribution technology had been deployed, it would send an alert to the Swedish Roadway Authority who would send out a gritter to reduce the ice on the road. If your car’s airbag was deployed in an accident, it would phone home and Volvo would potentially send out an ambulance if they couldn’t get hold of you on your mobile phone.
There’s a catch, though. A lot of that data collected in the process will not be useful. There aren’t enough changes between states (which is good for the driver), but over time insurance companies will be able to obtain a full understanding of each driver and provide a personalized insurance policy that balances the driver’s style with risk. And of course, even the catch has a catch: This data set is constantly growing and changing.
Thus, a vehicle telematics program that delivers real insights into the oncoming flood of data will need massive computing power in order to manage the data, determine which data is useful and which isn’t, and package and process it so that the insurance company can make high-quality business decision that meets both their needs and their customers’.
The Ultimate Solution for Car Telemetry Ingestion and Data Prediction
From a technical perspective, we believe the best way to manage this is a system that offers in-memory computing. In-Memory computing is the idea of utilizing RAM, which is basically the fastest storage medium on machines, as the primary storage, as well as business logic processing medium for any type of workflow, including event-driven data generation, as is the case with connected cars.
Here’s the good news: GigaSpaces’ InsightEdge platform was architected precisely to handle streaming events at the pace in which they are generated using in-memory real-time processing.
A live InsightEdge use case can be shown using Magic’s xpi. Predictive car maintenance requires car telemetry ingestion and data prediction. Magic’s solution stack needed one more component in the architecture to be fully compliant with fast data and scalable scenarios assured innovation was needed and the correct puzzle piece to fit. The flexibility of combining transactional and analytics functionality is what separates GigaSpaces from the rest. With Magic’s use case, we are enabling IoT applications at scale through open source components at the center, edge, and cloud.
If you’d like more information about how GigaSpaces can provide real-time insights, such as whether someone is a good, bad or even dangerous driver, contact us to schedule a demo.