If you already know you need to automate palletising, the next question is which kind. This page is the decision: the questions to answer in order, how the four options compare, and what a given set of answers usually points to.
If you are earlier than that and want to know how palletising cells actually work — the components, grippers, pallet patterns and infeed — start with automated palletising systems instead.
Five questions, in this order
The order matters. Rate narrows the field faster than anything else, and space is the one that most often overrules a preference formed earlier.
1. What rate does it have to keep up with?
Cases, bags or trays per minute — and specifically the peak, not the average. A line that runs comfortably most of the day but produces a burst after a changeover has to be palletised at the burst rate or buffered ahead of the palletiser.
This single number eliminates more options than any other. Very high rates rule out cobots entirely; modest rates make a layer palletiser hard to justify.
2. How many product formats will it handle?
Not how many it runs today — how many it will see over the life of the machine. Count case sizes, bag types, tray formats and anything seasonal.
One format points towards layer palletising. Several point towards a robot, where a new format is largely a programming and gripper question rather than a mechanical rebuild.
3. How often does the pattern change?
Distinct from format count. A line might run three products that all stack the same way, or one product with two different customer pallet specifications. Frequent pattern changes favour robotic systems, because changeover is a recipe selection rather than a mechanical adjustment.
4. How much floor space is genuinely available?
Measure the area you can actually give up, not the area on the drawing. A palletising cell needs the machine footprint plus the pallet positions, empty pallet storage, safe access for changeover and maintenance, and the guarded envelope around it.
Space is the constraint that most often overturns a decision made on rate alone, and it is worth establishing before anyone quotes.
5. Is the budget capital or incremental?
A single capital purchase and a staged programme lead to different answers. Where the appetite is to prove the concept first and expand later, a cobot cell or a semi-automatic step gets something working sooner and funds the next stage. Where the case is already made and the rate is high, staging costs more than it saves.
The four options compared
Relative rather than absolute — actual figures depend on product, pattern and gripper.
| Throughput | Footprint | Product flexibility | Changeover | Entry cost | |
|---|---|---|---|---|---|
| Layer palletiser | Highest | Large | Low | Mechanical, slower | High |
| Robotic arm | High | Moderate | High | Recipe change | Moderate to high |
| Cobot | Low to moderate | Small | Moderate | Recipe change | Lowest of the automated options |
| Semi-automatic | Operator paced | Smallest | High | None | Low |
Two things the table cannot show. The gripper often matters more than the machine — a robot that cannot reliably hold your product is not flexible in any useful sense. And every option depends on product arriving correctly oriented and spaced, which is upstream work regardless of what you choose. Both are covered in automated palletising systems.
What typical answers point to
Three illustrative combinations. They are worked examples of the logic rather than descriptions of specific installations.
One product, steady moderate rate, tight floor space
Say a single case format at around twenty cases a minute, in a facility with no spare floor area and no plans to add variants. Rate rules out semi-automatic as a long-term answer; space and cost rule out layer palletising.
That points to a compact robotic cell or a cobot palletiser, with the deciding factor being whether the rate sits inside what a cobot can sustain including pallet changes. The readiness signs are worth checking against.
Several formats, moderate rate, frequent changeovers
Four or five case sizes, a couple of pallet patterns per customer, and changeovers more than once a shift. Rate alone might suit layer palletising, but changeover frequency makes it the wrong tool — every mechanical adjustment is lost production.
This is the clearest case for a robotic arm, where a format change is a recipe selection. The real work moves to the gripper, which has to hold every format in the range.
High rate, one format, more than one line
A single product running fast, with two or three lines that could feed one palletising area. Cobots are out on rate; a robot per line is expensive and duplicated.
This is where layer palletising earns its cost, potentially as a multi-line installation with product from several lines merged and buffered ahead of it. The buffering is not optional — see accumulation.
Where the decision usually goes wrong
- Specifying to average rate. The peak is what the palletiser has to survive. A system sized to the average needs buffering it was never given.
- Counting today’s formats only. The variant added eighteen months later is the one that defeats the gripper.
- Treating space as a detail. More decisions are reversed by floor area than by any other single factor, and it is usually discovered late.
- Choosing the machine before the gripper. The end effector determines what can actually be handled. On difficult products it should be settled first, not last.
- Ignoring what happens after. A palletiser feeding a manual wrapping station moves the bottleneck rather than removing it.
Common questions
Can I start small and upgrade later?
Often, and it is a reasonable strategy where the case is not yet proven. What makes it work is designing the surrounding conveying, buffering and pallet handling for the eventual rate rather than the starting one, so the upgrade is a machine change rather than a rebuild.
How do I know if a cobot is fast enough?
Work from cycles rather than headline speed. Each case is a pick, a move and a place, and the pallet exchange stops production unless there is buffering or a second pallet position. Comparing your peak rate against realistic cycle time including those interruptions is the honest test.
What if we run several very different products?
That points to a robot, and the question becomes whether one gripper can handle the full range or whether a tool change is needed. Where a product is genuinely awkward, proving the handling before committing is worth the time — that is what proof of concept development is for.
Does a bigger machine future-proof the decision?
Rarely on its own. Over-specifying rate costs capital and floor space without addressing the constraints that actually change — product formats and pallet patterns. Flexibility usually future-proofs better than headroom.
How long does installation take?
The machine is rarely the long pole. Lead time on the palletiser, the conveying and pallet handling built around it, and the shutdown window available to install it generally set the timeline. A well-defined requirement shortens all three.
Do we need to decide before talking to a supplier?
No. But having the five answers above — peak rate, formats, pattern changes, floor area, budget shape — means a first conversation produces a proposal rather than a questionnaire.
Working it through
LVP Automation specifies and installs palletising systems for manufacturers in Ireland, including the conveying, buffering and pallet handling around them. The equipment detail is in automated palletising systems; the wider capability is on our automation services page.
We are based in Finglas, Dublin 11. Send us your peak rate, your product formats and your available floor area and we will tell you which option fits.
