THE INTRALOGISTICS CORNER

with Mike Nissen,
Commercial Director APAC at Hannover Fairs Australia/ Asia-Pacific

Intralogistics: An Incubator for AI and Robotics

A quick hello before we get moving

Welcome to the first edition of The Intralogistics Corner. Over the coming issues, this page will be a place to talk plainly about what is happening inside the four walls of our warehouses, distribution centres and fulfilment operations — the technology, the people, the economics, and occasionally the hype that needs a gentle deflating.

My role as Senior Commercial Director APAC at Hannover Fairs Australia/ Asia-Pacific comes with an unusual privilege. Every year I walk hundreds of exhibition stands, sit through countless technology demonstrations, view recently completed warehouses and talk with the engineers who build these systems, the integrators who install them, and the operations managers who have to live with them afterwards. I don’t sell robots and I don’t buy them — which means I can call things as I see them. That’s the spirit I’d like to bring to this column: a practitioner’s-eye view of the industry, minus the sales pitch.

So, for this first outing, let me start with the biggest thing I see happening: our industry has quietly become the place where the world’s AI and robotics grow up.

The perfect nursery

If you were designing the ideal environment for artificial intelligence and robotics to mature, you would end up designing a warehouse.

Think about what a young technology needs. It needs an environment structured enough to be manageable, but messy enough to be a real test. It needs a clear, measurable definition of success. It needs tolerant users who will work around its early failures. And it needs a commercial problem urgent enough that someone will pay for version one, not just admire it.

Intralogistics offers all four. A distribution centre is a semi-controlled world: no rain, no pedestrians pushing prams, no oncoming traffic — but plenty of genuine chaos in the form of damaged cartons, mislabelled pallets, seasonal peaks and the infinite variety of things humans manage to put in a tote. Success is ruthlessly measurable in picks per hour, order accuracy and cost per unit. And the commercial pressure is real: labour is scarce and expensive across nearly every market this magazine reaches, and consumer expectations of same-day delivery are not going backwards.

That is why so much of the AI and robotics we will all eventually meet in hospitals, airports and city streets is being incubated, right now, between the racking.

The evidence on the floor

The numbers tell part of the story. The warehouse robotics market is valued at around US$11 billion this year and is forecast to more than double to roughly US$24.5 billion by 2031 — a growth rate north of 17 per cent annually. What should interest LogiSYM readers most is where that growth is happening: Asia-Pacific already accounts for nearly 40 per cent of global revenue and is the fastest-growing region, with China alone recording a 44 per cent surge in new warehouse robot installations in a single year, and India now accelerating on the back of government automation incentives. This is not a trend arriving in our region from elsewhere. Our region is the incubator’s incubator.

But walk any CeMAT show floor and you see the qualitative story, which is more interesting than the market sizing. Autonomous mobile robots were the novelty act a decade ago; today the frontier has moved to orchestration — AI-driven fleet management that coordinates hundreds of AMRs like a flock of starlings, recalculating paths in real time and learning from yesterday’s bottlenecks. Computer vision, trained by deep learning, is quietly automating the unglamorous work of goods-in inspection, returns processing and quality control. The warehouse management system is evolving from a system of record into something closer to a co-pilot: prioritising orders, balancing workloads and handling exceptions before a human ever sees them. And digital twins now let operators rehearse a peak season, a supplier failure or a new layout entirely in software, before a single dollar is spent on steel.

None of these capabilities was invented for the warehouse alone. But the warehouse is where they became products rather than papers — because the warehouse could measure them, afford them and forgive them while they learned.

The humanoid question

No column on this topic in 2026 can dodge the humanoids. They dominated the conversation at the big international shows this year, with Chinese manufacturers well out in front and European integrators running early pilots on tasks like case picking and parts handling. The economics driving the interest are blunt: in parts of Europe, fully loaded labour costs now exceed €40 an hour, and younger workers understandably don’t queue up for repetitive manual roles.

My honest read? The incubator argument applies here more than anywhere. Humanoids will not earn their keep on public streets or in aged-care homes first — they will earn it, or fail to, in warehouses, doing bounded, repeatable tasks where a one-to-two-minute pick cycle can be timed, costed and compared against the alternatives. Today those cycle times are not yet competitive, and most credible manufacturers are talking about volume rollouts closer to 2030 than to next year. I’d also note the counterargument, put well by the traditional robotics camp: it doesn’t matter what a robot looks like; it matters what task it performs. A mobile base with an arm often does the job without the theatre of two legs.

Both things can be true. The humanoid form factor may or may not win — but the proving ground where we will find out has already been chosen, and it has racking in it.

What this means for you

If you run an operation, the incubator framing changes how you should think about adoption. You are not being asked to bet on science fiction; you are being offered technologies that have already survived their infancy in facilities much like yours. The sensible playbook I see among the better operators in our region is consistent: fix your data before you buy your robots, because AI is only ever as good as what it ingests; favour modular, phased deployments — shuttle systems and AMR fleets you can scale in increments — over big-bang projects; and design for hybrid operations, because the future on every serious show floor is people and machines dividing the work, not machines replacing the building’s entire workforce.

And if you build or integrate this technology, the message runs the other way: the intralogistics sector is the fastest feedback loop your product will ever get. Treat it that way.

Until next issue

The incubator never sleeps, and neither does this region — nearly every week brings a new facility, a new pilot, a new lesson learned the hard way. In future editions I’ll dig into individual corners of this world: the software layer where I believe the next real battle will be fought, the economics of automating brownfield sites, and what mid-sized operators — not just the retail giants — can realistically deploy today.

Until then, if you find yourself near an exhibition hall full of robots, come and say hello. I’ll be the one asking the demonstrator what happens when the carton is wet, the barcode is torn, and the picker is on a break. That’s where the real answers live — and it’s exactly where AI and robotics are learning their trade.