US Cold Storage’s AI leap: What it means for logistics

The AI agent achieved an 87% success rate in booking appointments and secured 96% accuracy

Alan operated around the clock and processed more than 150 scheduling requests simultaneously (Photo: Jim Allen/FreightWaves)
Gemini Sparkle

Key Takeaways:

  • US Cold Storage implemented FourKites' AI agent, "Alan," to automate and streamline the manual, time-consuming process of scheduling inbound and outbound delivery appointments across its 38 facilities.
  • During an 8-week pilot, Alan achieved an 87% success rate in booking appointments and 96% accuracy for requested delivery dates, resulting in a productivity boost equivalent to 36-40 human hours saved.
  • Unlike traditional automation, Alan utilizes context-aware AI to adapt to various communication methods (email, web, phone) and leverages FourKites' Intelligent Control Tower for broader supply chain awareness and adaptability.
  • This successful deployment transforms a major logistics bottleneck, enhancing US Cold Storage's operational efficiency, improving customer experience, and positioning the company for deeper AI integration and autonomous decision-making.
See a mistake? Contact us.

United States Cold Storage has introduced automated appointment scheduling through the use of FourKites’ AI agent, “Alan”. US Cold Storage, which operates 38 facilities nationwide, has long struggled with the complexities of coordinating inbound deliveries and outbound dispatch in a highly regulated, sensitive environment. 

The company identified appointment scheduling as one of the biggest efficiency drains; logistics teams spend hours juggling emails, phone calls, portal logins, and constant back-and-forth to secure delivery windows. Recognizing that automation could free staff to focus on higher-value work, US Cold Storage turned to FourKites.

Join the leaders shaping freight’s future at
F3: Future of Freight Festival, Oct 21-22.
Network with the industry’s best and discover what’s next.

Over an eight-week pilot, Alan demonstrated what many assumed was impossible. The AI agent achieved an 87% success rate in booking appointments and secured 96% accuracy in obtaining customers’ requested delivery dates. In handling over 600 shipments, the digital agent effectively compressed work that would have taken dozens of human hours into real time, estimating a productivity boost equivalent to 36–40 hours during the test period. Alan operated around the clock and processed more than 150 scheduling requests simultaneously, versus the sequential pace typical of human staff.

“The first individuals who interacted with the agent were actually amazed and very intrigued by this product and the idea of a digital agent and how—through utilizing technology—an appointment was just made with no human interaction,” said Keith Mowery, Executive Vice President, Logistics & West Region, United States Cold Storage. “The group was very excited by how easy it was to use.”

Alan differs from traditional robotic process automation (RPA) by relying less on rigid scripts and more on context-aware intelligence. Rather than requiring custom programming for each variation in retailer rules or portal quirks, it adapts in real time to email workflows, web interfaces, and phone interactions, all while maintaining audit trails and status visibility. 

This success builds on a long-standing relationship between US Cold Storage and FourKites, which began with shipment tracking and visibility services. The deployment highlights FourKites’ modular philosophy: start with one use case, validate value, then expand deeper AI capabilities across operations. For US Cold Storage, that means not just smarter scheduling, but the potential to layer autonomous decisioning across the full logistics stack.

FourKites frames Alan as more than an automation tool; it lives within its Intelligent Control Tower, a platform already processing over 3.2 million supply chain events daily. That network intelligence gives Alan awareness of the broader supply chain context, something standalone automation tools lack. 

In and beyond cold storage, scheduling appointments has long been a manual choke point across logistics industries, costing time and resources. But as Alan proves, agents grounded in real-time visibility and AI reasoning may be poised to take over the work once left to legions of coordinators.

For US Cold Storage, the reward isn’t just efficiency, but differentiation. With smoother operations and leaner staffing, the company is now positioned to elevate the customer experience while containing costs. And for FourKites, the success adds momentum to its vision of AI becoming a ubiquitous presence in supply chain control towers, not just as a back-end assistant but as a working partner in execution.

Mary O'Connell

Former pricing analyst, supply chain planner, and broker/dispatcher turned creator of the newsletter and podcast Check Call. Which gives insights into the world around 3PLs and Freight brokers. She will talk your ear off about anything and everything if you let her. Expertise in operations, LTL pricing and procurement, flatbed operations, dry van, tracking and tracing, reality tv shows and how to turn a stranger into your new best friend.