Kardex adds AI checks before AutoStore bins enter the grid

Kardex has launched BinInductAI to check and register containers while an AutoStore warehouse is being commissioned. Announced on 27 October, it automates a step before the storage grid begins routine work.[1]

An operator places a bin at the induction port. Computer vision checks its presence, walls, robot pickup holes and barcode readability. The system then asks AutoStore whether that identity is already registered, before sending the insertion commands.[1]

A container has its own identity

AutoStore’s documentation describes bins stacked beside and above one another inside a grid. Each has a unique number recorded in the controller’s database and repeated on a physical label. The controller knows where each container is; a robot can retrieve it and deliver it to a workstation.[2]

The contents belong to another record. AutoStore says the warehouse management system maintains the stock inside the bins, which can be divided into compartments for different products. Identifying a container therefore answers where the stock travels. It does not, by itself, establish what products or quantities have been put inside.[3]

Consider an illustrative bin containing two product types in separate compartments. Its container number can stay the same as workers remove one product and replenish the other. The stock record changes while the physical container remains identifiable. Keeping those records distinct allows the warehouse to follow both the reusable box and its changing contents.[2][3]

Where the time saving applies

Kardex claims up to 30% less on-site induction time. That concerns introducing bins during implementation, rather than subsequent order-picking speed. The release provides no trial size or detailed baseline for the figure.[1]

The new check combines physical inspection with a database lookup: a readable barcode still has to belong to a bin the system has not already registered.[1]

Report an error or suggest a correction