Anyone involved in transportation knows all too well that disruptions are a part of life. When they happen, the top priority is to find the safest, fastest way to keep freight moving.
At Amazon, the Relay Operations Center serves as the main point of contact for drivers experiencing delays. Agents are available day and night to provide immediate support. If a disruption happens, the agent must synthesize the many details of the issue and identify the best plan of action.
The task can quickly become complex for agents who need to sort through a host of information about the load: origin, location, destination, standard operating procedures, etc. They then need to determine the downstream impact of the issue and how it will affect future legs that the driver needs to complete.
Giving agents more power
This is the exact type of situation where Generative AI (GenAI) thrives. GenAI can process vast amounts of information from multiple sources in seconds. It learns patterns, draws conclusions, and generates new insights.
Amazon built a GenAI-driven app using AWS’s tools that can reduce agents’ cognitive load—the mental effort required to process information—and help them reach the most effective solution faster. The app uses natural language processing to quickly pull together essential information.
Relevant data points typically include prior disruption history, the trailer’s current location, lane, and transit times. Then, it generates easy-to-digest summaries so agents can swiftly identify the best next steps.
“Every innovation at Amazon starts with our customers in mind, including the drivers and agents who support them,” says Fernanda Rivas, a principal product manager who led the team that created the app. “When we built this capability, we focused on solving real problems, not just showcasing technology.”
For instance, imagine a driver is stuck in traffic. First, Amazon’s systems will detect signals to notice if the vehicle has stopped moving. The app will merge those signals with traffic information and determine it is tracking to be, say, an hour late. The app will guide the agent to contact the driver to check in on the status and get a better understanding of the traffic jam.
An hour late might not be a big deal for that current leg of the driver’s tour but, let’s say, the driver has two more stops where she is picking up and dropping off other trailers. That hour could have accumulating downstream effects. The app will crunch the data and give the agent recommendations of what to do – which could be to remove the next two legs from the driver’s tour so they can be picked up by other drivers and stay on time. It even calculates the pickup and delivery times for loads crossing time zones, which can be tricky, especially in the 15 U.S. states with multiple time zones.
This all empowers agents to get accurate answers in seconds and gives them greater control in gathering and analyzing information. Ultimately, the goal is to make agents’ jobs easier and ensure consistent, timely resolutions to get customers’ loads moving again.
Tapping into the tech
If you want to tap into the Amazon network that uses all this tech, you can do so through Amazon Freight. You get access to our 70,000+ trailers and Amazon’s advanced capabilities. Enterprise businesses should contact us here. Or, if you’re ready to try booking a load today, create an account.
