A.I. is getting your holiday gifts to you more efficiently than ever

The program optimizes three main areas of shipping logistics with the benefits of A.I.: use insights to make the FedEx network operate more efficiently; give customers more visibility and control over their supply chains; and solve e-commerce-related challenges.

It’s a big undertaking but one that Sriram Krishnasamy, chief executive officer of FedEx Dataworks, says is the direction that logistics companies lean today: using data-driven, digital insights to inform decision-making and increase transparency. In the first six months, the team launched a predictive insights-based sensor data for added visibility for packages, which he says had an immediate impact on vaccine distribution. At the time of publishing, the team has rolled out more than 40 additional solutions currently being used by FedEx and its partners.

“For most companies, the data generated across every aspect of the supply chain is something that they are simply trying to ‘manage’ rather than cultivate,” says Krishnasamy. “Data and information are far too often trapped in silos, tracked and reported on without having their full strength realized.”

As every aspect of commerce undergoes radical change thanks to technology, the supply chain has seen its cracks exposed, especially during the pandemic. The nuances of the complex system, which its sole goal is to transport goods from a manufacturer to an end customer, are susceptible to several variables, and the growing unpredictable nature of consumer behavior has challenged the process. Logistics companies like FedEx, as well as DHL and Amazon, have tapped into the potential of A.I. to help solve problems and get ahead of future hurdles. Why? The power of the technology to analyze billions of data points quickly provides useful insights for a logistics team to make educated decisions.

Kraig Foreman, president of e-commerce at DHL, says that the shift from brick-and-mortar retail with the exponential growth of e-commerce to online shopping has demanded operational growth from any company touching the supply chain. When customers predominantly shopped in-store, predictive modeling was much simpler. Stores knew, for example, that holiday shoppers would likely step inside after work hours or on the weekends, ensuring shelves were stocked at those times.

With e-commerce, that customer behavior went out the storefront window. Now, customers shop at any hour of the day, sometimes even placing orders back-to-back. That threw a wrench into the entire system, putting pressure on warehouses to have more stock at all times and shipping companies to figure out how to get goods to the paying customers as quickly as they historically picked them up from in-store shelves.

“You’d put it in your car and deliver the package home to yourself,” Foreman says of past consumers. “Now you’re expecting the supply-chain and logistics arm of that industry to do that for you: find that item, pick it, pack it, put it in a bag, check you out, and deliver it to your home on your behalf—the complexity in that.”

For the warehouse component, A.I. is critical to its success. Foreman says that models can be run to predict needs for orders to coincide with a marketing push (such an Instagram influencer partnership), and machine learning powers the strategy of assisted picking robots who source items in warehouses. The work being done today addresses how to add more inputs to daily forecasting models to better predict needs tomorrow outside of major manipulations like holidays and marketing campaigns.

At Amazon, A.I. predicts scenarios or limitations that would affect its “delivery promises”—or the date at which a customer expects to receive their order—including everything from how many seasonal associates are required in a metro area to minimum number of orders needed to meet demand. Adam Baker, vice president of global transportation at Amazon, says that some of the forecasting begins years in advance. Since adopting the tech in 2016, A.I. has shown inefficiencies in speed, cost, and carbon impact, equipping the Amazon team with necessary information to adapt and evolve to streamline the whole process.

A.I. has been particularly impactful at Amazon for its last-mile network—the partners and carriers that deliver packages from freight centers to the customer’s door. Routing is a primary recipient of the technology, with algorithms calculating optimal maps and routes that are safe and achievable for millions of deliveries each day, explains Beryl Tomay, the vice president of last-mile delivery tech at Amazon. She also says that A.I. has been integral to meeting sustainability goals by using environmental and physical considerations—including distance, attitude change, air temperature, and vehicle model—to predict energy consumption for each route. From there, the team adopted alternative delivery methods—including bicyclists, walkers, and electric Rivian vans—in more than 100 cities across the U.S., a project achieved in just three years thanks to the speed and efficiency of A.I. technology.

However, A.I. for shipping logistics remains in its infancy—and it’s hardly perfect. Many e-commerce customers can regale listeners with a story (or three) when a gift arrived late or when the online tracking provided the wrong date. Companies still grapple with reckoning sustainability initiatives with emissions from getting goods from point A to point B. And the infrastructure of American shipping systems doesn’t directly translate to technological innovations like electric vehicles. Courtney Muller, president and chief corporate development officer at Manifest, a logistics and supply-chain conference, notes that in 2023, 130 billion parcels are expected to be shipped around the world; by 2026, that number will double to 262 billion.

“That’s a short window for these shippers to figure out how to double their capacity for shipping,” Muller says. “When you start employing A.I. behind these really complex problems, you suddenly open up solutions no one thought of before.”