An FMCG robot is an autonomous machine — most often an autonomous forklift, pallet stacker or mobile lifting robot — that moves finished goods, raw materials and packaging around a fast-moving consumer goods plant or distribution centre without a driver on board. Every one of them lands on your risk register the day it is commissioned, because the Provision and Use of Work Equipment Regulations 1998 (PUWER) apply to autonomous trucks exactly as they apply to a manned counterbalance truck. For a UK FMCG procurement lead, though, the problem this quarter is rarely the machine. It is that three supplier quotes arrive in three different shapes — one bundles software licences into the unit price, one prices integration as a day rate, one hides charging infrastructure in a footnote — and none of them can be reduced to a cost per pallet move your capex committee will recognise, and the sign-off slips another quarter.
Why FMCG robot business cases stall in procurement
FMCG warehousing is a volume business run on thin margins. Promotional peaks arrive with four weeks' notice, SKU counts churn, and the pallet flow between line-end, bulk store and despatch cannot be allowed to queue. That makes the operational case obvious to everyone on site — and the commercial case hard to write, because the people who approve it are trained to compare like with like.
Procurement already holds good benchmarks for manned material handling equipment. A counterbalance truck on a five-year full-maintenance contract near Magna Park or DIRFT has a known annual cost, service inclusion and residual position. Nothing in a first robot proposal maps onto that. Behind the machine sit fleet control software, an integration to the systems the site already runs, charging infrastructure, floor and racking remediation, commissioning, and a safety case that must be evidenced rather than asserted.
The deeper problem is coupling. Where a proposal ties the machines to one manufacturer's control software, the first purchase quietly sets the terms of every purchase after it: tranche two stops being a tender and becomes a variation order. Committees spot that shape even when they cannot name it, and the paper goes back for rework. Regulation adds a third layer — the Health and Safety Executive (HSE) treats the equipment operator as duty holder, not the supplier, and workplace transport guidance applies wherever autonomous trucks, manned trucks and pedestrians share aisles.
PUWER 1998 applies to an autonomous forklift in an FMCG plant exactly as it applies to a manned counterbalance truck, from the day it is commissioned.
Lever 1 — Reprice the fleet as cost per pallet move, not cost per truck
This is the operational lever, and it unblocks the paper. Ask every supplier to restate their proposal against one denominator: the fully loaded cost of a single pallet move, over the full term, at your actual volumes. To do that you must count the moves. Take one normal week and one promotional week, and log every pallet movement by origin, destination, hour and distance — line-end to bulk store, bulk store to pick face, pick face to despatch lane, returns and rework.
That table does three things. It converts a machine specification into a throughput requirement, so you can see whether four robots or seven are proposed for the same job. It exposes the genuinely expensive hours: the third shift, the weekend replenishment run, the two hours after a line changeover. And it gives the committee a unit they already use, having signed off cost per case for years.
| Cost line | How it usually arrives | Ask for it as | Why the committee cares |
|---|---|---|---|
| Machines | Unit price per truck | Unit price plus assumed availability | Fleet size follows uptime, not headline capacity |
| Fleet control software | Bundled into unit price | Separate annual licence, renewal stated | A bundled licence hides the cost of tranche two |
| Integration | Day rate, scope to be confirmed | Fixed scope, named interface, ownership | Open-ended day rates are where projects overrun |
| Charging infrastructure | Footnote or exclusion | Line item with electrical and floor works | Capital spend on your building, not the robot |
| Floor and racking | Excluded | Survey, then a costed remediation schedule | Brownfield slabs decide if the fleet works at height |
| Service cover | Reactive call-out rates | Response times, spares, UK engineer cover | Availability claims are worthless without cover |
| Exit | Silent | Cost to redeploy or end the term | Footprints change inside seven years |
Lever 2 — Make the control layer a vendor-neutral requirement
This is the technical lever, and one clause here protects five years of spending. The machines are not the lock-in; the control layer above them is. Whoever holds the fleet manager decides which vehicles can join, how traffic is arbitrated, and what a different class of robot costs in three years.
Specify that the fleet management layer must command vehicles over an open standard rather than a proprietary interface. VDA 5050 is the relevant one — an open specification for communication between mobile robot fleets and a central fleet control, allowing one fleet manager to issue orders to vehicles from different manufacturers without a bespoke integration for each. Ask whether suppliers speak it natively or through an adapter, and ask for a live production reference.
Specify the business-system seam too. Your existing warehouse and enterprise systems should stay the record of stock; the fleet layer takes work from them and writes completions back. Ask what happens when the link drops mid-shift — a sound integration finishes the issued task, holds safely, queues completions and replays them without duplicating stock movements. FlyWei builds that seam with M4 fleet management and RDS robot dispatch, which is why a FlyWei fleet mixes machine types under one control stack.
Lever 3 — Specify PUWER, ISO 3691-4 and floor flatness up front
This is the regulatory lever, and leaving it to handover is the most expensive mistake on this list. Under PUWER your organisation is the duty holder for equipment provided for use at work, and a purchase order does not transfer that. Write the evidence into the specification: a risk assessment for mixed autonomous and pedestrian traffic, defined safety zones, documented emergency stop behaviour, and training records for supervisors who will run the fleet from a screen rather than a seat.
Name the standard as well. ISO 3691-4 covers driverless industrial trucks and their systems, and a competent supplier should already be working to it. Where machines lift, LOLER 1998 sits alongside PUWER; BSI publishes the British adoptions and Logistics UK is a practical route to sector guidance. Then survey the floor: autonomous trucks in high-bay racking tolerate slab deviation far less than a driver who compensates instinctively. Commission a TR34 flatness survey before the fleet size is fixed, so remediation sits in the same paper as the machines rather than surprising a site at Burton-on-Trent or Daventry.
Lever 4 — Match the commitment term to the contract you actually hold
This is the commercial lever. FMCG procurement rarely holds a site or a customer contract for as long as an outright purchase assumes. Buying outright pushes the whole cost into one year's capital budget and leaves you owning depreciating assets whose redeployment you never priced. Structuring the same fleet over a term converts it to operating expenditure, spreads it across the years that generate the savings, and keeps the exit visible. FlyWei offers warehouse robot leasing over 3, 5 and 7-year terms, so the commitment aligns to a customer contract or lease break rather than an accounting convention. Ask every supplier to price the same fleet all three ways — a committee that sees all three approves faster than one shown a single number.
What FlyWei does here
FlyWei is an independent, vendor-neutral UK systems integrator of autonomous forklifts and AMRs. We are not an OEM, a reseller or a distributor, which means we specify the best machine for each flow across multiple manufacturers rather than fitting your plant to one catalogue. For an FMCG procurement team that independence is the commercial point: it keeps tranche two a competitive tender instead of a variation order.
A FlyWei engagement starts with the pallet-move survey above, not a machine list. We count the moves, identify which are genuinely expensive, and size a fleet against your real shift pattern. Line-end clearance and bulk-store putaway point to FlyWei autonomous forklifts — counterbalanced classes for palletised cartons, narrow-aisle reach classes for high-bay. Tote and cart flows point to FlyWei lifting robots. Both run under one M4 fleet manager and RDS dispatch layer.
We then hand procurement what the committee asks for: a costed line-by-line proposal in the shape of the table above, a TR34 floor position, a PUWER and ISO 3691-4 compliance schedule, and the same fleet priced outright and over 3, 5 and 7-year terms. Comparable work is set out across our FMCG robot solutions by industry pages. Engineering and support are UK-based.
Frequently asked questions
What is an FMCG robot?
An FMCG robot is an autonomous machine — typically an autonomous forklift, pallet stacker or mobile lifting robot — that moves finished goods, packaging and raw materials around a fast-moving consumer goods plant or distribution centre without an on-board driver, directed by a fleet management layer that takes work from the site's existing warehouse system.
How should procurement compare FMCG robot quotes?
Restate every proposal as a fully loaded cost per pallet move across the full term, using your own measured volumes. Require machines, control licences, integration, charging, floor remediation, service cover and exit costs as separate line items. Quotes that cannot be broken down this way are not comparable.
Do we have to replace our WMS to run an FMCG robot fleet?
Usually not. An orchestration layer sits above the warehouse system and takes work from it, so that system stays the source of truth for stock and orders. Replacement is only worth considering where it cannot expose work through an interface or scheduled export — and even then, middleware is cheaper.
Does PUWER apply to driverless forklifts?
Yes. PUWER 1998 applies to work equipment provided for use at work, and an autonomous truck is work equipment. The site operator is the duty holder, not the supplier, so risk assessment, safe zones, maintenance and supervisor training remain your responsibility. ISO 3691-4 is the relevant safety standard.
Should we buy an FMCG robot fleet outright or lease it?
Lease terms suit most FMCG procurement teams because site tenure and customer contracts rarely match the life of an outright purchase. Leasing over 3, 5 or 7 years converts capital spend to operating expenditure and keeps redeployment and exit visible. Price the fleet all three ways and let the committee choose.
How do we avoid being locked into one robot manufacturer?
Make the control layer the requirement rather than the machines. Specify that the fleet manager must command vehicles over an open standard such as VDA 5050, ask whether that is native or via an adapter, and require a live production reference for a mixed-manufacturer fleet. A vendor-neutral integrator keeps later tranches competitive.
If a stalled FMCG robot business case is on your Q3 risk register, the fastest way forward is a costed fleet size built on your own pallet-move data rather than a supplier's assumptions.
Request a fleet-sizing and ROI estimate for your DC — we survey the moves, size the fleet to your real shift pattern, and price it outright and over 3, 5 and 7-year terms. You can also review current warehouse robot leasing terms first.
UK-based engineers. No obligation. We reply within one business day.