Skip to main content

Why Levi Strauss Built a Machine Learning-Powered Shipping Optimization Engine

Levi Strauss & Co. (LS&Co.) is hoping a new machine learning technology can help it circumvent some of the current issues plaguing its supply chain and distribution networks.

As part of the company’s digital transformation journey shooting for $10 billion in revenue by 2027, Levi Strauss developed the Better Optimization Of Shipping and Transport (BOOST) engine. The BOOST engine is designed to efficiently fill e-commerce fulfillment orders by using machine learning and artificial intelligence (AI) to identify the best location to dispatch each individual item.

“When somebody goes online to make a purchase, we have distribution centers where we keep inventory specifically for those orders,” Louis DiCesari, global head of data, analytics and AI at LS&Co., said in a statement. “One of the things that we can do with BOOST is broaden that search for available product.”

At the moment, BOOST accounts for nearly 40 percent of e-commerce orders, according to Inna Saboshchuk, LS&Co global operations data science manager.

“By Black Friday this year, we should be processing 100 percent of BOOST-eligible U.S. orders, and we are currently scaling this across our businesses in Europe,” Saboshchuk said.

Related Story

The company walked through how the technology works: A consumer who wants to purchase a Levi’s Trucker jacket online in a specific wash might found it’s out of stock. Typically, an out-of-stock product isn’t available in the company’s distribution centers. However, BOOST can determine whether the item is in a store near the customer and make it accessible to that shopper.

“We don’t want to miss the opportunity to delight shoppers and get them their next favorite Levi’s wardrobe staple because it’s not technically available to them online, but easily accessible in a local storefront,” DiCesari said. 

BOOST’s AI-based calculations allow LS&Co. to reduce its shipping impact without shifting the burden to the customer, the company said.

In addition to opening up more inventory to consumers, BOOST employs artificial intelligence capabilities to improve operational efficiency and cut costs for customers while still delivering a premium product.

Levi Strauss also said the system could even solve the problem of split shipments.

“If a consumer orders three things from Levi’s and gets three packages from Levi’s, it’s a poor experience for the customer and for the environment,” DiCesari said.

The engine is designed to consider all elements of the process, from shipping to packing to labor, optimizing decision making and saving time and money.

“The beauty of it is that we’ve been able to automate all of this so it’s really a decision-making engine, not just an information engine,” DiCesari said. 

The technology launch comes as the company continues to fortify its distribution capabilities. Last year, Levi’s Henderson, Nev. distribution center became the firm’s first owned-and-operated facility to fulfill orders for e-commerce, brick-and-mortar retail and wholesale channels.

Over a 10-month span, the team transformed more than 100,000 square feet of open space in the facility’s second floor into a warehouse that could serve the West Coast’s e-commerce business.

Now Levi Strauss is establishing a new e-commerce distribution center in Northern Kentucky slated to open in 2023. The company is investing more than $48 million and creating approximately 300 jobs at the existing 575,700-square-foot facility in Erlanger, Ky.

Overseas, the company broke ground on a LEED-certified 750,000-square-foot state-of-the-art distribution center in Dorsten, Germany that achieved WELL Health-Safety certification as well.

Aiming to begin operating in April 2024, the facility will manage the distribution of apparel, accessories and footwear across wholesale, retail, digital, e-commerce and marketplace channels. LS&Co. expects to employ up to 650 workers when the facility reaches full capacity of 55 million units in 2026.

LS&Co. has sought to blanket its employees with machine learning knowledge as it develops its data-driven aspirations. Last summer, it launched an eight-week, fully paid machine learning bootcamp to bring AI education to employees across departments, from teams in corporate to employees in stores and distribution centers.

“Fashion has been one of the biggest offenders when it comes to climate change,” said Katia Walsh, LS&Co.’s chief AI officer, during VentureBeat’s Transform 2021 virtual conference. “AI can save fashion because it can deliver us the sustainability, the creativity and the profitability that a company like ours aspires to have.”

For an institution that was founded in 1853, Levi Strauss & Co. has made it a point to evolve with the times. According to 2021 data from algorithmic merchandising firm Nextail, L&SCo. is the most “retail data-forward” of 22 major fashion companies researched, with a score of 27.4. This metric is measured by the total number of data-related professionals divided by the revenue of each brand studied.