The warehouse is full. Somebody has put a proposal on the table for a shuttle system, a conveyor, a set of mobile robots or a full automated store, with a payback figure at the bottom. Sitting next to it is a much duller quotation for reworking the racking and the layout.

We sell the duller one, so take the following with that in mind — but the honest position is that automation genuinely wins in some warehouses and genuinely loses in others, and the thing that decides it is rarely the technology. It is six facts about your building, your lease and your stock profile, and you can check all six before anybody quotes anything.

1. How long do you actually hold the building?

This is the question that gets left out of the business case most often, and in Singapore it is the one that decides more cases than any other.

Conventional racking is bolted to the floor. It comes apart, it moves, and it goes back up in the next unit — which is a substantial part of what you are buying, and it is the subject of relocating pallet racking to a new warehouse. A fixed automated installation is a different proposition: it is engineered to that building's grid, its floor and its clear height, and moving it can cost a serious fraction of buying it.

So put the remaining term of your lease, including whether you actually hold an option to renew and on what basis, next to the payback period in the proposal. If the payback runs past your tenure, the business case depends on a renewal you do not control. That does not kill the idea — but it changes it from an investment decision into a property decision, and it should be discussed as one. The tenancy angle on racking generally is covered in racking in a leased warehouse.

2. What is your stock profile, honestly?

Automation pays for throughput. Racking pays for density. Which one you need is a fact about your stock, not a preference.

If a small number of SKUs move constantly, in uniform unit loads, at a rate that is reasonably predictable, that is the profile automation was built for. If you hold many SKUs of which a modest proportion move in any given week — which describes a great many Singapore distributors, sub-contractors and trade suppliers — then most of your building is storage, not movement, and paying for throughput you do not use is the expensive mistake.

The number to establish first is simply how many pallet positions the building can hold and how many you need, which is worked through in how many pallet positions your warehouse holds.

3. Are your loads uniform enough to be handled by a machine?

A machine handles what it was designed to handle and stops for everything else. Uniform pallets of a consistent size, weight and condition are the easy case. Mixed pallet sizes, overhanging loads, timber crates, drums, coils and anything shrink-wrapped by hand to an inconsistent shape are the hard case, and the exceptions do not disappear — they get handled manually, off to one side, which is where the labour saving quietly goes.

If you already know that mixed pallet sizes are a live issue in your operation, that is a signal in itself; the conventional side of that problem is set out in mixing pallet sizes in selective racking.

4. What does your floor say?

This is the one that catches people out after the order is placed.

Conventional selective racking is tolerant of an ordinary industrial floor. As the equipment gets taller and more precise, the floor tolerance tightens sharply — this is already true of very narrow aisle operation, and it is more true of automated systems. Automated installations also tend to concentrate load into fewer, heavier points.

Two surveys settle it: floor flatness and floor loading. Neither is expensive relative to what is being proposed, and both should happen before an order, not after. Warehouses have discovered a flatness problem after the equipment arrived, and the fix — grinding or re-screeding a floor in an operating unit — is worse than any of the alternatives. The flatness question is set out in the floor flatness reality of VNA conversions, and floor loading in a flatted or upper-floor unit has its own constraints, covered in flatted factory racking and floor loading.

5. How much height can you legally use?

Automation buys density mostly by going up. If the building's clear height is modest, or if sprinkler clearance caps your top beam well below the roof, the density advantage shrinks — and you may be paying a large premium for a gain that a taller conventional configuration would also have delivered.

Measure the true usable height before you compare density claims: not roof height, but clear height to the lowest obstruction, less the clearance the fire protection design requires. The measuring is in warehouse clear height and racking levels, and the clearance rule that governs the top of the rack is in sprinkler clearance and racking approval.

6. What happens on the day it stops?

When a conventional rack is hit by a forklift, you derate or isolate one bay, put a barrier round it, and the other ninety-nine bays keep working while a replacement upright is sourced. The failure is local and the recovery is understood — that is the whole logic of repair, replace or derate.

When an automated aisle stops, that aisle stops, and everything stored in it is unavailable until it is running again. That is not an argument against automation — it is an argument for asking three specific questions before signing:

• Who holds spares in Singapore, and which parts are held locally rather than shipped?
• What is the contracted response time, and what does it cost?
• Can stock be retrieved manually while the system is down, and how long does that take?

A vendor with good answers to those three is a different proposition from one who has not been asked.

Before you spend anything: the layout you already have

The cheapest capacity in most warehouses is not new equipment. It is the configuration of what is already installed, and it is routinely worth a double-digit percentage of positions:

• Beam pitch. Levels set for the tallest pallet in the building, everywhere, wastes vertical space in every bay. Setting pitch to actual load heights is the single most common gain.
• Aisle width. Aisles are frequently wider than the trucks require, because they were set once and never revisited. The relationship is in forklift turning radius and aisle width.
• Slotting. Fast movers at the wrong end of the building, or on levels people have to reach into by hand, cost time every day and create a safety problem as well - see picking from high racking levels.
• Going up before going wide. Where clear height allows, adding a mezzanine or additional levels usually beats every other option per dollar; the arithmetic is in mezzanine versus warehouse expansion.

Do that first. If the building is still short of space afterwards, the automation case is a real one and you will be making it on clean numbers instead of on a layout nobody has revisited in six years.

When automation is the right answer

To be fair to the other side of the argument, there are clear cases:

• High, steady throughput on uniform unit loads. The profile the technology was designed for.
• You own the building, or hold long secure tenure. The payback question disappears.
• Cold store. Where every cubic metre is expensive to keep cold and hours of manual work in a chilled space are hard to staff, the case changes completely - see what changes in a cold room.
• Labour at height is the real constraint, not floor space. If the problem you are solving is people working at height rather than storage density, that is a different and better reason.

The short version

• Put the payback period next to the remaining lease term. In Singapore that comparison decides more cases than the technology does.
• Automation pays for throughput; racking pays for density. Check which one your stock profile actually needs.
• Mixed and irregular loads become manual exceptions, and that is where the labour saving goes.
• Get floor flatness and floor loading surveyed before an order, not after.
• Measure true usable height after sprinkler clearance before comparing density claims.
• Ask who holds spares locally, the contracted response time, and whether stock can be retrieved manually.
• Fix beam pitch, aisle width and slotting first. That capacity is nearly free.

If you want an independent read on whether your building has capacity left in it before you commit to anything, WhatsApp us on 9107 2601 — tell us the site, the clear height, roughly how many pallets move a day and how much lease term is left, and we will tell you honestly whether a layout rework gets you there. What we install and how it is designed is on our pallet racking and services pages.