Skip to Content
Artificial intelligence

How machine learning is accelerating last-mile, and last-meter, delivery

Wise Systems’ routing software helped one company cut late deliveries by 85%.
March 25, 2019

While much of the logistics industry’s efforts to accelerate delivery times focuses on optimizing routes, it turns out that’s not where drivers spend most of their time.

In fact, as much as 75% of their workday is dedicated to navigating not the “last mile” but the last 100 meters—waiting at loading docks, searching for parking, and interacting with customers, said Chazz Sims, chief executive of Wise Systems, a startup based in Cambridge, Massachusetts, that has developed autonomous routing and dispatch software.

Using data and machine-learning tools, the company found that this kind of service time varies widely depending on the time of day, the specific customer, the goods in question, and the delivery person, Sims added. For instance, certain shops get busy serving customers at particular times of the day, or receiving goods from different delivery trucks at others. By spotting those patterns and shifting schedules around, the company was able to cut down delivery times and costs.

Wise Systems’ tools automatically adjust routes, drivers, and schedules throughout the day in response to other shifting conditions as well, including weather, traffic, and backed-up loading docks. By analyzing data from the fleet of brewing company Anheuser-Busch, one of the startup’s biggest customers, Wise Systems was able to cut late deliveries by 85% and fleet miles by 13%.

The business, founded in 2014, raised $7 million from Google’s AI fund late last year.

Deep Dive

Artificial intelligence

What does GPT-3 “know” about me? 

Large language models are trained on troves of personal data hoovered from the internet. So I wanted to know: What does it have on me?

An AI that can design new proteins could help unlock new cures and materials 

The machine-learning tool could help researchers discover entirely new proteins not yet known to science.

DeepMind’s new chatbot uses Google searches plus humans to give better answers

The lab trained a chatbot to learn from human feedback and search the internet for information to support its claims.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.