Layer 6 — Case handling mechanics and storage geometry

These are the hard physical bounds of any proposed solution. Reach, payload, depth and spacing decide what a layout can do; no amount of software cleverness moves them, and every layout discussion comes back to these numbers.

ACR mechanism variants

What it is. ACR — Automated Case-handling Robot — is Hai's own product category: a mobile robot that carries multiple cases (totes/cartons) and retrieves them from standard racking. Hai ships five mechanisms.

Why it matters here. Choosing the mechanism is a large part of solution design. Each is a different answer to "how do you get a case off a shelf?"

Fundamentals — the competitive landscape you must be able to place them in.

The ACR proposition in one sentence: most of the density of an AS/RS, on racking you may already own, deployable incrementally.

Hai's five mechanisms:

Plus HaiClimber (powering HaiPick Climb) and two AMR classes, K50 (fast-transit) and K600/K1000 (heavy-duty).

How Hai applies it.

"Innovative hook extracting design, bringing higher safety and efficiency." — /robots/haipick-a42-e6s

"The lifting fork can handle materials of different regular shapes, such as material trays, totes, and carriers" — /robots/haipick-a3

"has been specially designed to handle totes stored on single- up to triple-deep racks, and its maximum payload can reach up to 300 kg" — /robots/haipick-a42-e6s

Tradeoffs and pitfalls. Mixing models raises spares and training cost, yet Hai's own deployments do exactly that (32 A42 + 8 A42T) — because the telescopic units serve the high racking and the cheaper units serve the rest. That is a sensible cost-optimisation to be able to explain.

Key terms. ACR · AS/RS · AMR · AGV · goods-to-person · shuttle · cube storage · telescopic mast · grappling hook · double-/triple-deep.

Multi-tote batch handling

What it is. One robot carries up to nine cases per trip.

Why it matters here. This is the economic core of the ACR concept, and the number that makes the throughput arithmetic work.

Fundamentals.

• Retrieval cost decomposes as: travel to aisle + vertical lift + extract + travel to station + deliver. Travel dominates. Amortising one round trip across 9 cases divides the per-case travel cost by up to 9.
• The gain is only realised if the 9 cases are wanted at roughly the same time — which is precisely what order batching upstream provides. Hardware capacity and software batching are the same design decision seen from two ends.
• Rough throughput model. With cycle time T (round trip), k cases per trip and N robots at utilisation u: cases/hour ≈ 3600 × N × k × u / T. Sanity-check with a real figure: the UK cross-border site reports 1,500 cases/h.
• Little's law (L = λW) is the companion tool: totes in flight = arrival rate × time in system. Useful for sizing buffers and queue depth at workstations.

How Hai applies it.

"HaiPick ACRs can batch deliver up to 9 containers in a single run, reducing travel time and maximizing operational efficiency." — /solutions/system-features

"System picking efficiency average 1,500 cases/h" — /cases/uk-cross-border-ecommerce-warehouse-project

Tradeoffs and pitfalls. Carrying 9 cases only helps if batching finds 9 useful ones. With a low-affinity order profile (single-line orders, huge SKU count) the effective batch collapses and so does the business case.

Key terms. cycle time · batch size · utilisation · Little's law · picks per presentation · amortised travel.

Multi-depth storage and aisle economics

What it is. Storing totes one, two or three deep in a rack.

Why it matters here. Depth is the primary lever on storage density, and it trades directly against retrieval complexity. Being able to reason about that trade is core solutions-architect work.

Fundamentals.

• Deeper storage means fewer aisles for the same footprint. Aisles are pure overhead — the classic single-deep layout can be 30–40% aisle.
• The cost is digging: retrieving a buried tote requires relocating the ones in front. The relocation rate depends on how well slotting groups items by velocity and how often a buried item is wanted.
• Rule of thumb: put slow movers deep, fast movers shallow. This makes depth and slotting the same decision.
• Mixed-depth layouts are the sophisticated answer: some aisles double-deep for the A/B items, some triple-deep for long-tail stock.

How Hai applies it.

"Reduce aisle area by 50% and further increase storage density." — /robots/haipick-a42

"HaiPick Robots can work with aisles having different depths, e.g., double-deep racks in some aisles and triple-deep in others." — /solutions/haipick-system-3

"Supporting single to triple-deep tote storage options, making the best of every inch of usable space" — /robots/haipick-a42-e6s

Mixed-depth support is explicitly stated — that is the flexible option, and worth designing around.

Tradeoffs and pitfalls. Density claims quoted without a relocation-rate assumption are meaningless. Ask what order profile the density figure assumes.

Key terms. single/double/triple-deep · digging · relocation rate · aisle ratio · volumetric utilisation · mixed-depth layout.

Case spacing and density engineering

What it is. Winning storage density in millimetres of clearance.

Why it matters here. It shows where the real engineering effort goes, and it is the clearest illustration that density is a mechanical tolerance problem.

Fundamentals.

• Storage density in a tote system = (tote volume) / (tote volume + clearances + structure). Clearances exist to absorb positioning error. Reduce positioning error and you can reduce clearance, which converts directly into locations.
• Compounding matters: saving 20 mm per case across thousands of positions is metres of rack, which is aisles, which is building.
• This is why the mechanical-constraint trick in navigation pays twice — it removes a sensing step and tightens tolerance.

How Hai applies it.

"Horizontal case spacing up to 30mm* while reducing the gap between front and back of totes to 0mm." — /robots/haipick-a42-e6s

"The size of the cases can be automatically recognized for flexible pick and place." — /robots/haipick-a42

"Utilizing case spacing to increase the overall storage density by more than 30%." — /robots/haipick-a42

Flexible-width forks recognise case size and adapt — spacing becomes dynamic per case rather than fixed to the largest.

Tradeoffs and pitfalls. Tight spacing reduces tolerance for rack deflection, floor settlement and damaged totes. Density bought at the tolerance limit becomes a reliability problem two years in.

Key terms. clearance · positioning accuracy · tolerance stack-up · rack deflection · flexible-width fork · volumetric density.

Container envelope and payload limits

What it is. The hard physical bounds: what fits, how heavy, how high, how fast.

Why it matters here. These are the first numbers you check against a customer's SKU profile. If their cases fall outside the envelope, no amount of software helps.

Fundamentals.

• Always check the customer's case dimension distribution, not their average. The 5% of oversized SKUs decide whether you need a second handling method.
• Height drives density but also cycle time — vertical travel is slow, so the top of a 12 m rack is the cheapest storage and the most expensive retrieval. That is precisely why slotting puts slow movers there.

Hai's published envelope:

How Hai applies it.

"HaiPick System 3 easily accommodates containers ranging from 300×300 mm up to 850×650 mm." — /solutions/haipick-system-3

"ACRs can transport containers within these dimensions" — /solutions/system-features

"Extreme speed of up to 4 meters per second, enabling fast order fulfillment and extended cutoff times" — /solutions/system-features

"The standard load is 30 kg. The maximum load can be customized to 50 kg." — /robots/haipick-a42-e6s

⚠️ These figures are scattered across marketing prose — there is no spec sheet on the site. The downloadable brochures that would carry full specifications are form-gated. Treat the table above as indicative and verify before quoting numbers at people who know them precisely.

Tradeoffs and pitfalls. "Up to 4 m/s" is a top speed, not an average; real cycle times include acceleration, deceleration, turning and queueing.

Key terms. envelope · payload · reach height · duty cycle · case dimension distribution · top speed vs. average speed.

Container and racking agnosticism

What it is. Running on standard racking and handling many container types rather than requiring proprietary structure.

Why it matters here. It is Hai's central competitive differentiator against grid/cube systems, and it changes the shape of a proposal: lower capex, faster deployment, incremental scaling.

Fundamentals.

• Proprietary-structure systems (grids, fixed shuttles) achieve higher density but demand a purpose-built installation, a large up-front commitment, and a site that will not change.
• Standard-racking systems trade some density for retrofit capability and incremental scaling — start with one aisle, grow with the business.
• The strategic argument: for a customer whose volume is uncertain, optionality is worth real money. Deploy into part of an existing building, expand if it works.

How Hai applies it.

"Most HaiPick Systems can operate with various types of racking, not requiring a specific type." — /solutions/system-features

"The system can handle totes, cartons, and trays simultaneously in one warehouse." — /solutions/haipick-system-3

"Users can re-use existing racks and totes if they fit all requirements." — /robots/haipick-a42

"Our robotic warehouse automation system can help you rapidly automate your warehouse without impacting any running operations." — /robots/haipick-a42

Note the qualifiers — most systems, if they fit all requirements. And recall that the grappling-hook model needs a guard board fitted to the rack, which is a genuine tension with pure agnosticism. Knowing both sides of that is the mark of someone who read carefully.

Tradeoffs and pitfalls. "Any racking" claims meet reality in rack tolerance, upright spacing and floor flatness. Site survey exists for a reason.

Key terms. retrofit · brownfield vs. greenfield · incremental scaling · proprietary structure · site survey · optionality.