By April Durrett, Global Head of Experience, Verizon Connect
August 14, 2024
April Durrett, Verizon Connect
AI is everywhere. It’s in the media. It’s pop culture. And now, executives are clamoring for it. Sales wants it, as does the C-suite. I’m going to let you in on a little secret, though: AI is not a silver bullet. Sure, it’s a powerful tool, maybe even a game-changing tool, but it’s still a tool. You don’t lead with it, looking for problems it can solve. That’s backwards. You always start with the challenge and build around it. AI can enhance solutions, but it doesn’t change the central tenet of fleet management: start with the needs.
It can be difficult to make a value judgment on a driver’s performance. Part of the challenge lies in the prospect of a driver getting into trouble and having their position in the company threatened. Some fleet managers are partial to leaderboards, believing that they might encourage friendly competition and, in turn, elevate the standard of driving across the fleet. In my experience, however, this approach creates friction among drivers. Those ahead in the standings become boastful and may even put down other drivers lower on the totem pole. Needless to say, this is not conducive to improving driver performance.
The leaderboard is a clear example of coming up with a solution before understanding the problem. That solution starts with an assumption instead of a question. The right question is: How do I lower risk among my drivers and my business? By asking this question, you’ll discover that drivers often need help to better understand their shortcomings and guidance to improve their skills and behaviors behind the wheel. Video coaching is extremely effective in supporting driver development.
An effective solution requires understanding both the drivers’ needs and challenges, as well as the macro fleet challenges. Fleets and their drivers contend with a host of costly and dangerous safety issues, many of which are context-sensitive and dynamic. For example, some streets are narrower in certain cities, while roads in others may be hiller or curvier. Some cities are more congested than others, and different geographical locations are subject to different weather and environmental conditions–all of which have an impact on safety, fleet risk, and driver performance.
Contextual coaching sessions with drivers can help them better navigate all of these variables. AI can enhance this feature by combing through large tracts of data to identify context-specific driver tendencies and patterns and help human coaches support their drivers in a quick and productive way. The combination of dashboard cameras, vehicular sensors, computer vision, and AI can generate a driver profile that identifies areas for improvement. For instance, a driver may have a tendency to accelerate and decelerate too abruptly in urban areas. This not only poses safety issues and raises insurance considerations but also consumes more fuel. One sub-performing driver won’t upset a fleet, but if dozens or hundreds of drivers display these tendencies, it will have a discernible impact on the fleet’s bottom line.
Having this context allows fleet managers not only to develop profiles for drivers but also to identify coaching opportunities based on specific events. For example, if a driver comes to an abrupt stop and nearly rear-ends a vehicle in front of them, a dash cam with AI can determine that the driver wasn’t looking at the road but rather at a phone on their lap at the time of the event. Meanwhile, sensors on the vehicle can determine how quickly the driver had to stop and how close they came to a collision. A central fleet management system may then be alerted to the event, providing the fleet manager with a specific, teachable moment. In addition, this context can help exonerate drivers involved in an accident who were driving safely. The opportunity is further strengthened if the driver’s profile shows a tendency to accelerate and decelerate faster than the average fleet driver.
Video coaching is an excellent example of how disparate tools can come together to achieve synergy. Neither dash cams, vehicular sensors, nor AI alone can create an end-to-end workflow to coach drivers based on behavior. Even with all of these tools working in tandem, a human must engage with the data and the driver for everything to work. Additionally, you must first understand the challenges drivers are facing and the kind of help they need to perform job functions optimally before creating a solution that can support those efforts.
AI is a powerful tool, but don’t rush to implement it. Make sure it dovetails with what you’re trying to achieve. Don’t worry about being first to market–focus on being best in market.
