12 October 2022
3 min read

Improving the efficiency and relevance of driver training with pattern AI

Individual driver risk profiles form the basis for targeted, contextual training

The wide-ranging benefits of driver training are being realized on a broader scale than ever before. As operating costs continue to rise, and the need to go green remains in sharp focus, driver training is being recognized as a valuable solution. While first and foremost it can help companies reduce collisions, a solid driver training program has many additional benefits, including helping to drive down operating costs and reducing environmental impact.

Driver training does have its challenges for companies, however, including costs, administrative effort, and driver downtime. Fortunately, the evolution of technology and Artificial Intelligence (AI) means that fleet operators have the tools at hand to deliver training to those who need it most.

Defining individual risk profiles enables training to be tailored to the specific weaknesses of drivers and delivered when it’s needed. Pattern AI’s real-time risk analysis ensures that fleets reap the benefits associated with delivering driver training, without spending time and effort on training that’s unnecessary or irrelevant.

Proactively identifying high-risk drivers

Across the fleet industry – and among the driving population in general – a minority of drivers are responsible for a disproportionate number of crashes. This is approximately 15% of drivers who cause 50% of crashes. Identifying the highest-risk drivers is important, as it enables training efforts to be concentrated on those drivers who are costing a company the most.

An important point to note is that risk status can constantly evolve, so a driver might move from low-risk to high-risk over time. There are many factors that can contribute to a change in risk level, including variations in driving routes, times/days of travel, and other work-related or personal factors. For this reason, it’s beneficial to measure risk on a continuous basis, rather than periodically.

Another consideration for many fleets is that vehicles might be used by several drivers. Pinpointing the driver risk level – as opposed to the vehicle risk level – is an important factor in ensuring training is delivered to those who need it.

Individual risk profiles for greater understanding

The method used to generate a driver risk profile is important. Checking a driver’s license history/motor vehicle record (MVR) is a commonly used method to measure driver risk. But such checks are reactive and might be conducted as little as once per year.

Some companies use other methods to measure driver risk, for example telematics. However, telematics is also a reactive approach since it relies on events – such as harsh braking, accelerating, or speeding – having already happened. There is also little evidence to support a significant connection between events and collisions, as discussed in a previous blog.

With the latest in pattern AI, companies can obtain an individual risk and CO2 profile for every driver based on real-time driving performance.

Pattern AI looks for similarities in driving behavior

Greater Than’s pattern AI works in a similar way to facial recognition. In the same way that facial recognition compares a person’s facial signature to a database of known faces, Greater Than’s pattern AI compares a person’s DriverDNA to a database of real-life trips.

Basic telematics roughly analyzes 5% of every hour, approximately 150 data sets. Greater Than’s pattern profiling measures approximately 3.6 billion data sets per hour. It’s easy to see, therefore, just how much deeper AI can delve into driver data to provide comprehensive, valuable, risk insights.

It’s easy to get started

GPS data is all that’s needed to get started with Greater Than’s AI – and this can be obtained via API, SDK, or a smartphone app connection. Many companies already have access to the data required for an individual driver risk profile, they just don’t have the AI that’s needed to provide deeper insights into risk and CO2 emissions.

Once a company connects to Greater Than’s cloud-based AI platform, our Enerfy database processes and analyzes every trip against billions of previous real-world trips. This identifies patterns in driving behavior and provides insight into crash probability per driver, relative crash cost and climate impact. The analyzed data can then be pushed to a company’s existing fleet management system, a driver app, SDK or visualized via Greater Than’s Risk Portfolio Tracker. It’s then up to the fleet itself to decide how to utilize the data and provide the training it feels is most appropriate for the identified risk level.

Whether you’re keen to access individual driver risk profiles to predict and prevent collisions, reduce fuel and CO2 emissions, or simply ensure your driver training program is targeted accordingly, we can help. Contact Greater Than or book a meeting with me to discover how your business could benefit from individual driver risk profiling.

Recommended resources

Who is owning the motor insurance industry revolution?

Who is owning the motor insurance industry revolution?

The prospect of a motor insurance revolution is a hot topic and there is currently much discussion around how technology is finally transforming the industry. But it’s not just motor insurance companies themselves that are part of the revolution, as other major players discover that by using their GPS data for risk analysis, they become part of the next generation of motor insurance influencers and providers.

Read more