Convoy Is Revolutionizing Trucking Through Machine Learning

Helping truckers and the environment

Every year, truckers in the United States cover more than 95 billion highway miles—enough to travel the globe over 3.7 million times. According to Convoy, a Seattle-based logistics company, nearly $800 billion will be spent on trucking services in 2018, moving 10.5 billion tons of cargo.


Trucking is, in short, a massive industry. But not necessarily an efficient one.

A staggering 40 percent of the miles truck drivers log each year are done with an empty truck, representing a costly waste of time and fuel. A large part of the problem is the industry’s infrastructure—a fragmented network of shippers and haulers, big and small, joined together by brokers that match one side with the other. This process often relies on traditional methods like email, address books and phone calls.

Convoy is disrupting the model by using artificial intelligence (AI) to automate it. “We’ve created a digital online marketplace through our mobile application where carriers and drivers can use it to directly find work,” says David Tsai, senior manager of marketplace and data platform engineering at Convoy.

Convoy’s approach uses machine learning, an AI technique, to provide better matches for shippers and truckers, allowing them to move freight more efficiently—and lowering costs for both parties—using Convoy’s matching system. Bigger shippers that have computerized systems in-house can also integrate Convoy’s online digital marketplace into their own.

Another benefit of the system is transparency. With Convoy, carriers can see the price offered for any job and make an informed decision that makes sense for them. On the flip side, shippers get instant price quotes so they can make comparisons between carriers.

"Leveraging AI to build out models to facilitate that relevance is something we put quite a bit of emphasis on.”

Casey Olives
Head of Data Science
Convoy

"Leveraging AI to build out models to facilitate that relevance is something we put quite a bit of emphasis on.”

Casey Olives
Head of Data Science
Convoy

Using Amazon SageMaker, Convoy’s machine learning models analyze millions of shipping jobs along with trucker availability, then recommend matches that are cost-efficient and timely. This impacts everything from routing and prices quoted to shippers and truckers, to recognizing what types of loads match best with individual drivers.

“When users log into their Convoy app, they’re able to look at a list of offers, and the ones at the top of the list are those most relevant to them and their business,” says Casey Olives, head of data science at Convoy. “Leveraging AI to build out models to facilitate that relevance is something we put quite a bit of emphasis on.”

So if a carrier has a job from Seattle to Los Angeles, for example, the app will even recommend work for the trip back. Reducing the miles that empty trucks drive is good for truckers—as well as the environment.

Amazon SageMaker allows Convoy to accelerate innovation and disruption in the industry. Previously, Convoy’s data scientists would create models and then hand them over to engineers to rewrite into production-level code. With SageMaker, this translation step has been removed. Data scientists now have the freedom to build machine learning models quickly, reducing their reliance on engineers.

“It has enabled us to iterate much more quickly, and actually move from development to deployment with much quicker velocity,” Olives says. “It’s making for a fast handoff between our data scientists and engineering.”

As Convoy works with more shippers and drivers, its AI can take advantage of more data from the entire freight network for demand forecasting. In other words, it’s a dynamic process—one that might finally bring efficiency to one of the world’s largest industries.

“As we work with more shippers and carriers, we get a better understanding of how much capacity is available and how much demand is coming in on specific lanes,” Olives says. “Being able to have a contextual view of the entire network will enable us to drive efficiencies in utilization and costs benefiting both carriers and shippers.”

Coinbase uses ML to create a secure cryptocurrency exchange

Learn more »

Capital One uses ML to better protect customers from fraud

Learn more »

Zendesk helps companies provide faster customer service using ML

Learn more »

T-Mobile humanizes customer service using ML

Learn more »