Until May 2025, Net Zero Logistics vans were operating 30 to 40 routes daily within Connecticut. Mark Chiusano, CEO and owner of Net Zero, which focuses on the last mile of delivery, said he wants more efficient routes that can meet customer needs while reducing costs.
Net Zero’s own traffic management software was unable to suggest optimized routes, so Chiusano said they decided to try Finmile, an AI-powered traffic routing software that provides dynamic routing based on location, weather, traffic conditions, service level agreements, vehicle specifications, driver behavior and available drivers. Net Zero now operates an average of 16 to 20 routes each day, with each driver delivering more packages in the same amount of time, Chiusano said.
Willem Jan van Hove, professor of operations research at Carnegie Mellon University’s Tepper School of Business, said mathematicians and traveling salesmen have been trying to make routes more efficient for centuries. “Travelling salesmen go from city to city to sell products. They wanted to sell in the most cost-effective way,” he said. the goal? Find the shortest distance to all stops. But today, the problem is even more complex.
“Look at the urban environment: rush hours, school closures, events. Even distance is highly dependent on travel time,” Van Hove says. He said the classic approach of prioritizing the shortest distance is not well-suited for delivery today, as customers expect delivery at a specific time.
Rich Pleeth, CEO and co-founder of Finmile, understands this modern problem first-hand through his experience in physical logistics. His company was founded in 2022 and delivered packages using electric cargo bikes and electric vans. Then, in 2024, they pivoted to route optimization software. “We realized that shipping was a broken part of e-commerce,” Preece said.
The company uses agent AI to capture orders, build initial routes, and continually update them throughout the day to improve efficiency. “AI doesn’t sit on a dashboard waiting for you. It actively makes decisions, reassigning stops if a driver is late, inserting turns into the actual route, proactively predicting delivery failures, and automatically initiating customer communications,” Preece said.
More efficient routes and baggage sorting
Traditional software systems often layer AI on top of existing code, which “can become unwieldy,” said Stuart Heiden, president and COO of Net Zero. But AI is the foundation of Finmile, making it easy for Heiden and his team to create customized reports and make routing changes without technical expertise, he said.
Route automation software has also helped Net Zero reduce the time drivers spend sorting packages. This process often begins after 3 a.m. when we receive the customer’s shipping information for the day.
Previously, drivers had to sort packages by geographic area, which was time-consuming and labor-intensive. “Dynamic routing makes it very fast and efficient. You scan a package and it shows up on your device and tells you which route to take,” Heiden said. “This will lead to a reduction in labor costs for sorting.”
Once Net Zero operations staff accepts the first AI-generated route suggested by Finmile’s algorithm each morning, the route is made available on the driver’s smartphone. Drivers can scan packages and load them onto the truck. The program continues to aggregate data and refine its algorithms to further improve routes.
“Our team has the ability to edit routes on the fly,” Chiusano said. “It actually appears to be reducing travel time and increasing the efficiency of each route.”
With Finmile, drivers no longer have to sort packages into delivery order. Instead, drivers place their luggage in their assigned totes. At each stop, software that the driver accesses on their device designates a tote number. Heiden said it’s faster for drivers to rummage through a tote bag of 30 to 40 items than to sort them by stop before hitting the road each morning.
Improved tracking and fewer customer complaints
Mr Chiusano said that since introducing FinnMiles, net-zero routes have become increasingly efficient and drivers are making more deliveries.
He added that AI-powered technology has helped reduce the number of unclaimed deliveries by giving drivers access to precise drop-off coordinates. In addition to geo-tag delivery via Finmile’s software, a barometer on each driver’s cell phone tracks steps. Preece said the program knows whether the driver is going upstairs or delivering on the ground floor, and that information is automatically fed into Finnmiles’ algorithm.
Similar to Amazon’s drivers, people using Finmile can take a photo of their drop-off location, including the address number and door. The software maintains a library of door colors and provides additional checks against future incorrect deliveries. Combining these AI-powered proof of delivery checks has reduced delivery claims, Preece said.
Dispatchers and drivers can also track their performance in real-time. The system displays the number of orders delivered by the driver, average delivery time, and driver dwell time. Giving drivers access to this information increases their transparency into performance-related management discussions, Heiden said.
Companies not using dynamic route optimization need to “be proactive,” Chiusano said. “The world of logistics is not the same as it was 10 or even two years ago.”