• AI can support last-mile delivery by optimizing truck routes and predicting errors before they occur.
• AI-enhanced predictive analytics can also help to prevent package theft.
• This article is part of "How AI Is Changing Everything: Supply Chain," a series on innovations in logistics.

In the last mile, the part of the supply chain that involves transporting goods from a warehouse to a consumer's home, many things could go wrong. A package could end up at the wrong address, shipments could be late due to traffic, or a thunderstorm could damage a parcel left out in the rain.

"You're dealing with humans and the real world and trucks and traffic," said Fred Cook, the cofounder and chief technology officer of last-mile delivery company Veho.

In an area long dominated by carriers like UPS, FedEx, and the US Postal Service, Veho and many other software providers are looking to solve the challenges that pervade this notoriously complex and expensive part of the supply chain. They're using AI to design more efficient delivery routes, improve accuracy and the customer experience, and predict errors before they might happen.

Erik Mattson, a partner in consulting firm AlixPartners' Manufacturing and Operations practice, sees "a big opportunity for AI to help this industry catch up to other industries."

E-commerce sales continue to grow, reaching new highs of $300 billion in the last two quarters. This makes the last mile busier than ever and ripe for a technology disruption. A McKinsey report found that in the last decade, about $80 billion in venture capital went to logistics startups, with on-demand last-mile delivery platforms getting the greatest share of those funds.

AI from the road to the front door

Last-mile routes typically involve multiple stops and individual small packages — rather than one truck delivering pallets to a single warehouse — making this supply chain segment difficult to manage efficiently and expensive for the businesses involved. Last-mile delivery makes up an estimated 41% of all logistics costs in the supply chain, according to the Capgemini Research Institute.

One of the earlier applications of routing technology in the last mile was a machine-learning application that UPS launched in 2013 called ORION, or On-Road Integrated Optimization and Navigation. Four years ago, the parcel company rolled out an upgrade to ORION, which shortened routes by an average of two to four miles per driver and rerouted drivers based on changing conditions.

"Historic technologies would be static and run the night before," Mattson said. If orders changed or construction started, the tech wouldn't account for those changes.

Today's AI models, on the other hand, adjust in real time.

"Compared to pre-AI methods that relied on static routing rules or dispatcher intuition, our platform now responds dynamically to real-world conditions at scale," said Andrew Leone, the CEO and cofounder of Dispatch, a last-mile delivery platform.

Dispatch uses AI to plan routes based on factors such as traffic, delivery windows, estimated time per stop, and driver capacity. More efficient routes can lower fuel costs, improve density, and enable more deliveries in a day, increasing revenue for providers.

Amazon has been at the forefront of bringing AI into its last mile, said Jett McCandless, the founder and CEO of project44, a supply chain software platform. Last month, Amazon announced an initiative called Wellspring, which uses generative AI to analyze satellite images, apartment building layouts, street imagery, consumer instructions, and photos from past deliveries. It can recommend which parking spot or apartment building entrance a driver should use to drop off a shipment. In a test this past fall, the tech identified parking spots at 4 million home addresses.

Veho uses AI for quality assurance on its deliveries. In an ideal world, Cook said, an employee dedicated to quality assurance tasks would look at the geocode of where a parcel was left, examine the delivery photo, gather feedback from the driver, and determine if anything should change for future deliveries.

"It's totally infeasible to do that on millions of deliveries. But those are the types of use cases that we see, in the very near term, that AI is ideal for," Cook said.

Delivery data also allows last-mile providers to keep consumers informed. Deliveright, a last-mile delivery service, saw customer service calls drop by 80% due to real-time tracking and more accurate ETAs, according to Doug Ladden, Deliveright's CEO.

Veho said that its large language model, which it created in-house, answers 60% of customer and driver questions and has cut average response times from 2.5 minutes to 15 seconds.

Predicting and preventing package mishaps

Veho uses AI to pinpoint commonalities among mishaps that occurred during the logistics process, like the same warehouse associate handling multiple packages that resulted in errors, or a trucking company in the middle mile that had damaged items.

The company forecasts the likelihood of issues for specific routes or deliveries. Then it makes decisions based on the patterns, like moving packages to different facilities or increasing rates on a certain route, so drivers will be incentivized to pick them up earlier in the day.

"We've taken that a step further now to where we're trying to predict defects," Cook said.

Swiped packages are a big issue in last-mile deliveries, with 58 million parcels stolen from doorsteps last year amounting to $16 billion in losses, according to a USPS watchdog report.

UPS created an AI-based software, DeliveryDefense, that analyzes historic factors such as loss frequency and delivery attempts. The AI then spots areas that could be targets for porch pirates in the future.

McCandless said AI can predict high-risk areas and times of day, allowing companies to plan delivery schedules and routes accordingly to minimize the chance a package might be stolen.

"AI could play a key role in identifying patterns, helping to prevent theft before it happens," McCandless said.

• Quick-service and fast-food restaurants typically collect data on customers' purchasing behaviors.
• With the help of AI, they can now leverage their data to better manage inventory and operations.
• This article is part of "How AI Is Changing Everything: Supply Chain," a series on innovations in logistics.

Fast-food chain Juici Patties, which operates more than 70 locations in Florida, New York, and Jamaica, started on the island nation as a family kitchen in 1978. When the chain expanded into the US last year, it experienced stockouts.

Executives knew they needed a different strategy — one with advanced technology to scale their business, manage franchises, and sell thousands of patties each day, Stuart Levy, the company's chief technology officer, told Business Insider.

Today, Juici Patties uses AI's predictive and proactive features to prevent disruptions before they occur.

"AI is helping to keep our distribution centers stocked with enough of our branded packaging to meet demand," Levy said.

Indeed, AI technology is making its way into quick-service and fast-casual restaurant operations. AI can use data to form predictions about customer orders, then generate insights for leaders on how to manage inventory and operations.

Domino's Pizza and Microsoft teamed up to create a generative-AI assistant that saves managers time on inventory management and ingredient ordering. Starbucks also inked a deal with Microsoft to use genAI in its product development. And Yum Brands, the parent company of KFC, Taco Bell, and others, partnered with Nvidia on AI for internal tasks such as labor management and analytics processing.

For many quick-service restaurants, "their entire brand is built on speed and efficiency," said Spencer Michiel, the restaurant technology advisor at Back of House, a resource for restaurant tech solutions. "If there's anything that can help them with speed, efficiency, and lower cost, they're going to jump all over it."

Data-rich restaurants layer on AI

Restaurants are "extremely data-rich," Michiel said, which makes them well-suited to adopt AI. Major fast-food chains already have standard operating procedures to purchase based on demand, but AI takes that to the next level with forecasting abilities that more accurately predict demand and inform supply.

With AI's forecasting capabilities, restaurants can predict what customers might order and use this data to buy ingredients, a notoriously tricky part of restaurant supply chain management.

"The biggest thing that restaurants do badly is purchase," said Stephen Zagor, a consultant focused on restaurants and food businesses and an adjunct assistant professor of business at Columbia Business School.

AI draws from quick-service restaurants' internal point-of-sale data, such as sales trends and which products customers tend to buy at the same time. Then, an AI algorithm combines this data with external factors like the weather or local events.

"The beauty of AI is it's taking forecasted demand and turning that into a reaction all the way through the supply chain," Zagor said.

For example, AI can deliver granular data by location. For a restaurant right off an interstate, AI could predict that travel will slow down on certain days. Seeing that prediction, restaurant managers could decide to drop their inventory levels and purchase fewer items, Zagor said.

He named McDonald's as one quick-service restaurant that uses AI to maximize everything from its point-of-sale to its supply chain. The fast-food giant has partnered with Google Cloud and IBM on various AI solutions.

When it comes to data and AI, the level of standardization across major chains puts them at an advantage over smaller franchises and independent restaurants.

A mom-and-pop restaurant may not have "the time, the bandwidth, the skills, the knowledge" to gather data and create an action plan, Michiel said. Subscribing to software can cost hundreds of dollars each month, presenting financial barriers to small businesses. Any new back-of-house or supply chain software would need to integrate with existing point-of-sale systems. If done incorrectly, the result could be data loss or lag, "and it's going to be frustrating," Michiel said.

Serving up efficiency and financial gains

AI's predictive power can also help minimize waste in restaurant supply chains. If a restaurant orders too much, it could have to discard unused or expired food. This could require the business to increase meal costs to compensate for the loss, according to Michiel.

"Food waste is just a killer," Michiel said. "Over-ordering is straight loss. There's no way you're going to recover that cost."

Controlling costs is especially critical for fast-food chains, which order at scale and sell low-priced products. Making just 5 cents more on an item, or making 5 cents fewer, "is a big deal," Zagor said.

AI can also promote cost savings by flagging if a particular ingredient swap could result in higher profits without sacrificing taste or quality. The technology "smooths out" a restaurant's ability to purchase inventory while still keeping customer satisfaction top of mind, Zagor said.

"You can get good profit, and the customer is going to be happy," Zagor said. "It's win-win."

Levy said Juici Patties' AI implementation into its point-of-sale system and supply chain was time-consuming, involved some growing pains, and sparked fears about replacing the workforce with AI. He acknowledged that "AI isn't flawless."

Now that the technology is in place, though, Juici Patties has seen a boost in operational efficiency, Levy said. In one instance, the AI revealed that customers wanted to purchase food earlier in the day, before Juici Patties locations were open.

"We were missing potential sales during earlier hours of the day," Levy said. The restaurant chain acted upon that information and adjusted its opening times. The result: "a consistent increase in daily sales," Levy said.

• Retailers like Target and Walmart use AI to prevent stockouts and manage inventory.
• AI systems can predict demand to help improve inventory accuracy and availability.
• This article is part of "How AI Is Changing Everything: Supply Chain," a series on innovations in logistics.

The adage "too much, too little, just right" isn't just for Goldilocks and her porridge. Balance is also critical in inventory management, the part of the supply chain responsible for analyzing what consumers will buy and making sure products are in stock at the right place and the right time.

Excess inventory can lead to markdowns or expired goods, but too little product can lead to shortages that impact a retailer's brand image, customer satisfaction, and bottom line.

To prevent inventory from running out, big-box retailers such as Target, The Home Depot, and Walmart are using AI to predict when product amounts could dwindle. As a result, Target's inventory availability has improved every year for the last four years, Prat Vemana, the executive vice president and chief information and product officer at Target, told Business Insider.

AI can help retailers proactively adjust stock before disruption strikes rather than reacting to changing conditions, said Ajoy Krishnamoorthy, the CEO of inventory-management platform Cin7. That's especially critical today, Krishnamoorthy added, when factors like consumer behavior, inflation, and trade policy constantly impact supply chains.

"AI thrives in this environment," Krishnamoorthy said.

From old methods to AI

Traditionally, companies procure inventory, manage logistics, and analyze consumer behavior in silos, said Vidya Mani, an associate professor of business administration focused on technology and operations management at the University of Virginia. Teams do individual research, then come together to develop a strategy and execute it.

"We no longer have that time," Mani said. "By the time you finish doing it, the world will have changed on you."

Before Target started using AI to predict stockouts, the retailer relied on software-based applications, which didn't react or adapt to real-world changes as quickly as AI systems, according to Vemana. In fact, Target said in a blog post that it previously failed to catch half of its products that were out of stock because the technology they used thought that inventory existed when it actually didn't.

Target changed how it managed inventory in 2023 with the introduction of Inventory Ledger, an internal tech system that tracks inventory changes across stores and uses AI to predict when products might be out of stock "even before it's obvious to team members or systems," Vemana said.

Today, Target uses AI-driven inventory management for more than 40% of its product assortment, which is more than double when it started two years ago, Vemana said.

With Inventory Ledger, algorithms pull in data like supply lead times, transportation costs, current inventory, and consumer demand. Some models are more accurate for frequently purchased categories, and others are better suited to discretionary purchases or clearance items, so Target uses both kinds. There are also models that detect items that are in the store but in the wrong aisle or shelf, Vemana said.

Target has a demand forecasting tool that "makes billions of predictions each week about how many units of each item we'll need in stores and online," Vemana said. Together, all of these technologies guide employees' decisions about when, where, and how to reorder products and replenish stock.

"Combining traditional software with AI helps us make smarter, faster decisions about inventory management and keep our stores stocked more consistently," Vemana said.

One of Cin7's customers, ABC School Supplies, is also using AI to access real-time sales data, potential stock shortages, and supplier lead times, so it can "reorder proactively and avoid costly gaps in supply," Krishnamoorthy said.

The AI-driven inventory management marks a big change from what ABC School Supplies did in the past. It used to copy and paste website orders into its system, make a pick-and-pack list on paper, walk that physical list over to the warehouse, and manually update inventory, Krishnamoorthy said

The Home Depot is also taking an AI-based approach to inventory. In 2023, the retailer rolled out a machine-learning-powered app called Sidekick, which guides store workers to restock shelves and find products on overhead shelves, among other features.

"It helps make sure products are on shelves for our customers, and it manages our on-hand accuracy, which feeds to our replenishment and selling platforms," a spokesperson for The Home Depot said.

AI's predictive power for inventory planning

Krishnamoorthy said that in retail, "AI is exploding" as businesses move away from static planning to dynamic forecasts that anticipate demand and prevent stockouts.

AI allows businesses to get more granular with their inventory data, avoiding stockouts at particular store locations or during peak times. Mani gave the example of cosmetics and how different stores will have varying product needs based on consumer demographics.

"AI can figure out those patterns of baskets that are bought frequently in these different clusters," Mani said. "You don't need to feed it that contextual information."

Target is also working on technology that predicts which colors and sizes of seasonal items will sell, so it can stock those items in specific stores "to meet local demand," Vemana said.

At Walmart, AI-based inventory management systems use algorithms to make sure stores in warmer states have the right amount of pool toys and colder states stock enough sweaters, according to a press release on the company's website. If a particular item isn't selling on the East Coast but it's flying off shelves in the Midwest, algorithms flag that pattern so Walmart can reposition its inventory.

As retailers continue to develop and deploy AI tools, Mani predicts the technology's use cases will advance over the next decade.

Mani said that in two to three years, AI will likely flag stockouts without a person needing to walk into a store to confirm. Five years out, AI could automatically reorder inventory when algorithms detect that stock is running low. And in 10 years, AI would understand how macroeconomic events (like inflation) will change future consumer purchase behavior patterns and adjust inventory plans accordingly.

"It will feed it into your rulemaking rather than reacting to the situation," Mani said. "You'll be living in a different world."