Every VersaTile system is driven by a single AI engine that re-solves the tile field continuously, in real time. It decides what moves, where and when — concurrently, across thousands of cells — without any pre-planning, batch windows or fixed routing. It's the reason the system adapts to what the operation needs right now, not what was scheduled for it.
Conventional automation plans a sequence and executes it. VersaTile doesn't plan ahead — it solves continuously. The engine evaluates the entire field at once and moves every cell concurrently towards the current objective, re-solving as priorities shift. Demand spikes, order changes and reprioritisation are absorbed in real time rather than queued for the next planning cycle.
Each tile serves the needs of the controller. Add tiles and the controller adapts, without having to reprogramme the system — capacity and intelligence scale together.
The engine knows where every item is and where it's going, right now — not aggregated, delayed, or batched.
There is no central controller doing the thinking. Every cell is controlled locally and in real time. When a tile stalls, the field re-solves around it and keeps running, with no single point that stops the operation. That is what lets the field scale without a bottleneck and report on the operation at the level of an individual SKU/container.
The engine learns over time. Patterns in demand, sequencing and presentation feed back into how the field is solved, so performance improves continuously from day one — without reconfiguration, downtime or a change programme. The longer it runs, the better it gets.
Demand patterns, sequencing and presentation behaviour all feed back into how the field is solved.
Improvement happens automatically — no change programme, no downtime, no engineer visit required.
There's no tuning phase. The engine works from the first live order.
Matrix and Grid run on a common AI engine and software platform. The intelligence doesn't care whether it's moving totes in three dimensions or pallets in two — the same concurrent, real-time model applies to both. Learn it once, integrate it once, run both products on it.
Matrix (totes) and Grid (pallets & cages) share much of the same AI engine, API, and software platform.