How is predictive data shaping the auto industry

Analytics   |   
Published March 14, 2017   |   

Predictive data is quickly changing the way that we work in a variety of industries, and the automotive world is no exception. From connected car production to inventory management, the collection and study of large amounts of data are quickly shaping how our cars are built. How is predictive data changing the automotive industry and what changes can we expect to see in the future?


Connected and autonomous cars are going to benefit most from the inclusion of predictive data because their design centers on data collection and processing. Teslas, for example, are constantly connected. They send and receive data about their location, local driving conditions, and a variety of other variables. This, when paired with machine learning algorithms, has allowed the auto-pilot program to ‘learn’ to become a better driver.
It’s not just limited to the experiences of one car either.  If one Tesla learns a new driving technique or a road hazard to avoid,  the rest of the Tesla’s on the road learn it as well. This improves the overall safety and efficiency of the entire Tesla program rather than just one car.


We’re all at least marginally aware of our car’s maintenance schedule. For decades, the parts that need to be changed on your vehicle have been based less on wear and more on the number of miles you’ve driven the car.  While it is a relatively accurate benchmark, predictive data could change how we look at automotive maintenance by predicting what we need to repair or replace before it ever becomes a problem.
This particular style of predictive data will use data from things like warranty repairs or common problems that keep occurring. A technician looking at a handful of cars might not see a pattern, but when a predictive data algorithm has the information from hundreds or thousands of cars across the country to work with, it’s easier to find a trend.
That information could save both customers and manufacturers money in the long run, by preventing the need for emergency repairs or in the case of the manufacturers, the need for expensive recalls.


As more and more connected cars hit the roads, data management is going to become an essential tool. Connected cars generate around 25 gigabytes of data per hour.  For an average driver, that’s just under seven terabytes of data per year. If all the cars on the road right now (around 253 million) were connected, we’d be generating more than 1700 exabytes of data every year.
Globally, manufacturers expect to have somewhere around 2 billion cars connected to the cloud by 2035. With those many cars connected, we would be sending somewhere around 14 zettabytes of data to the cloud every year.  This is just automotive data and doesn’t take into account the data generated by videos, music, and other sources.
New techniques for data management will be essential to process all of this data.  On the other hand, this massive data cache is ideal for the application of predictive analytics.  It could easily change how we handle all of our data, no matter what the source is.


For connected cars, cybersecurity is essential. This has already been demonstrated by the hack that was able to take control of and finally stop a new Jeep in 2015. While security is always high on everyone’s list of priorities, security flaws like the one that allowed the Jeep hack is a distinct possibility that has to be addressed.
This can and should start at the manufacturing stage.  Connected cars should not be designed in such a way that a hacker could take over control of the primary controls of the car, i.e. steering, engine, transmission, or brakes.  Brakes could be the one exception to this rule,  as being able to remotely stop the vehicle can be useful in the event of car theft.
Software for these connected cars should be tested for flaws before the cars ever leave the factory. Manufacturers should also pay attention to alerts sent to them by hackers who aren’t on their staff.  Working with the white hats can make it easier to identify potential security flaws before they become full-blown security problems.
Predictive data has already shown potential for preventative maintenance, but this same application could be used to predict software problems and security flaws as well.
Connected cars are the biggest innovation in automotive technology in the last few decades, so it’s expected that we hit a few bumps in the road before it really gets off the ground. Once it does get moving, the connected car could potentially help reduce traffic accidents and fatalities, as well as provide a solution to the high levels of traffic congestion in heavily populated areas.
This is, of course, all speculation but based on the impact that predictive data has had thus far on the automotive industry; we’re excited to see what comes next.