The SF AppWorks Blog

What I Learned About Trucking, Empty Miles, and Elastic Capacity

Written by Andrew Greenstein | Dec 8, 2022 8:23:02 AM

Trucking is an $800 billion backbone industry in the U.S. Seventy percent of all goods shipped domestically are shipped on a truck. When the trucking and freight network hums along seamlessly, we barely notice it. We all get our stuff. Shippers get paid. Truckers do, too. But when things go wrong — an extreme weather event, labor shortages, a global pandemic that stresses the supply chain — everyone feels it. 

 

On The Next Great Thing podcast, I talked with Dorothy Li, Chief Technology Officer at Convoy, about how technology is transforming trucking, a traditionally fragmented, paper-based, manual industry. Convoy is building a digital freight network using AI, machine learning, automation, and data analytics to connect truckers to shippers. It essentially acts like a digital broker or a virtual carrier, like the Uber of trucking. Truckers use an app that helps them optimize freight routes and get paid fast, while shippers benefit from greater visibility into their shipments, lower shipping costs, and expanded elastic capacity for freight — just like cloud computing did for compute capacity.

 

Check out the episode on our website, Apple Podcasts, Spotify, or anywhere you listen to podcasts.

 

Check out all podcast episodes.

 

One “great thing” I learned: Inefficiency is one of the world’s biggest polluters. And there’s a big opportunity to create more efficiency in trucking…or any other freight or transportation network. The U.S. transportation sector, which includes trucking, is our nation’s largest greenhouse gas emitter, accounting for about 27% of all total greenhouse gas emissions. I was shocked to learn that a lot of this pollution is avoidable! Every year, up to 35% of miles driven by trucks are “empty miles” — trucks driving around without carrying any freight. All those empty miles are releasing 87M metric tons of carbon emissions annually. 

 

Related: 11 BEST ENTREPRENEUR PODCASTS FOR 2023 - LESSONS FROM STARTUP LEADERS

 

Get a Free Copy of 'A Step-by-step Delivery of Custom Apps'

 

“One of the biggest problems in trucking that people don't really think about is that there are a lot of empty miles. Meaning that, if you have a route from Seattle to San Diego, that might be a shipment from one shipper, but you need to go home,” Dorothy says. “When you go home, how do you batch loads that allow you to also carry shipments on your way home? That often requires a combination of route planning, and that's where machine learning really comes into play. ” 

 

Making even a minor improvement in efficiency (in Convoy’s case, optimizing routes) can create a massive improvement across the freight network and, consequently, reduce carbon emissions. That’s why Dorothy and her team are solving for the classic “traveling salesman problem” in computer science, which seeks to find the shortest, most efficient route for data to travel between various points. In the trucking industry’s case, it’s finding and optimizing the shortest, most efficient route for trucks to travel between different cities. 

 

Route optimization is a complex problem in logistics, and one that the human mind alone can’t solve. But software and artificial intelligence can. “My machine learning scientist was telling me that the combination of routes sometimes can be 1065. It's more than the number of atoms combined!” Dorothy says. “That's what machines do really well. We use machine learning to do that route optimization.”


The benefits of route optimization software not across the logistics industry -- from trucking to rail to air travel to shipping -- are many. Lower shipping costs, improved efficiency, greater visibility into the supply chain, maximized resources, and better for the environment. 

 

What are some of the other “empty miles” in industries other than trucking and logistics that create unnecessary waste? How can this problem be solved through software that can create greater efficiency and elastic capacity? Let me know in the comments.