AI Is the Execution Layer
Inside the AI Transformation of Parts, People & Processes
Hi, I’m Lily. I live in the world of distribution, where the execution layer, the daily flood of forecasts, check calls, and routing decisions, has always been the hardest part to scale. That layer is still there. But AI isn’t sitting beside it anymore. It is it. From demand planning to appointment scheduling to warehouse design, the systems doing the actual work of distribution are now AI systems.
If you only read one thing this week, it is this:
Hormel Foods went live with o9 across 70+ distribution sites, replacing disconnected planning tools with an AI/ML system that drives touchless demand forecasting, AI-recommended inventory transfers, and optimized truckload grouping. Hormel didn’t add AI to its planning process. It replaced the process with AI.
What’s Working in the Field
Agentic AI Rewires the Modern TMS
The transportation management system has long been where humans review options and make decisions. Agentic AI is changing that model. Where a traditional TMS surfaces recommendations, an agentic layer carries them out: rerouting shipments, adjusting carrier assignments, flagging and resolving exceptions without waiting for a dispatcher to act. Shipwell argues this shift is already underway, as the dispatcher’s role isn’t disappearing, but rather moving up the stack, from execution to oversight.
Gartner: AI-First Warehouses by 2030
Gartner released analysis predicting that 50% of new warehouses built in developed markets will be designed as robot-centric facilities by 2030. The projection reflects a structural shift already visible in distribution: persistent e-commerce volumes are pushing companies to build for robots first. Facilities designed around autonomous mobile robots operate at a different cost baseline from day one, and this distinction is what makes this a desirable outcome to achieve by the end of this decade. Gartner’s forecast is about architectural decisions being made now, for facilities that open in 2026 and beyond.
Hormel Goes Live on AI Planning
Hormel Foods completed a multi-year planning overhaul, going live with o9’s AI/ML platform across more than 70 dry and refrigerated distribution sites in five sequential go-lives through 2025. The system replaced a collection of disconnected tools with a single enterprise model linking demand signals directly to supply, inventory, and deployment decisions. AI now drives touchless demand forecasting, reducing manual overrides, models seasonal demand drivers, recommends inventory transfers between locations, and groups truckloads by weight, volume, and stackability.
What’s In My Ears
AI and the New Trade Brarrier with Dravida Seetharam and Sarah Lahti
In this episode of Talking Supply Chain, Dravida Seetharam, Fellow at the Center for Global Enterprise, and Sarah Lahti of the Digital Supply Chain Institute examine whether AI is creating a new structural divide in global supply chains. Worth listening for: the argument that AI’s real barrier isn’t the technology, it’s the infrastructure behind it, including energy, capital, and data access, and what that means for operators competing across global markets.
Lily’s Quick Take
Look at this week’s stories as one frame, not three headlines. Hormel’s planning system isn’t advising; it’s deciding. Shipwell’s agentic TMS layer isn’t surfacing options; it’s closing them. Gartner’s warehouses of 2030 aren’t being designed to support workers; they’re designed to operate without them. The execution layer, the part of operations that used to require constant human attention to run, is now AI’s domain. The question isn’t whether to bring AI in. It’s which parts of your execution layer you’re ready to hand over, and which ones need a human in the loop.
Until next week—keep your systems learning!
— Lily @ InstaLILY AI
Thank you for reading! Have feedback? Email me, Lily, directly. I read every email.
LinkedIn • Twitter (X) • Web
