The trade press has spent this year calling 2026 the year of the pallet shuttle, and the reasons operators give are consistent: it scales a lane at a time, it retrofits into racking that already exists, and it does not hang the whole building off a single crane. Logistics Business, reporting on the current wave of shuttle projects, notes that buying criteria have widened past labour replacement to include system safety, scalability and lifecycle serviceability. Materials Handling World makes the same point from the operator side — shuttles are being chosen precisely because they avoid the single point of failure of a traditional AS/RS.
That resilience is real, but it is bought with maintenance. A shuttle fleet is a lot of small, mobile machines running on batteries inside a steel structure, and every one of them has a wear schedule. This is the checklist we would want on the wall of a shuttle-served facility: what to look at, at which interval, and why. The specific intervals and tolerances belong to your OEM manual — stow, Movu, Swisslog, Mecalux, Jungheinrich and the rest all specify their own — so treat this as the structure, and the manual as the numbers.
Every shift (operator-level, five minutes)
These are the checks that catch a fault while it is still a reset rather than a callout. They need no tools and no engineer, which is exactly why they get skipped unless someone owns them.
- Battery state of charge on every shuttle at shift start, and whether any shuttle hit its low-charge threshold earlier than the shift plan assumed. An early low-battery reading is the first symptom of a cell that is on its way out, and in cold storage it is often the first symptom of a battery that was never specified for the aisle temperature.
- Fault-log review — not just "were there faults" but "were there repeats". A positioning correction that ran twice on the same lane, or the same sensor error clearing itself three times, is a pattern, and patterns are what preventive maintenance exists to catch.
- Visual check of the lane entrances and the lift/transfer station: debris on the rails, broken pallet boards, stretch-wrap tails hanging into the channel. Most "mysterious" shuttle stops are a piece of a pallet in the wrong place.
- E-stop and light-curtain function at the lift station if your OEM specifies a shift-start test. If they do, it is a safety-critical check and belongs in the record with a name against it.
Weekly
Weekly checks are the first ones that need someone with a torch and a tool bag rather than a scanner. They are about the interface between the shuttle and the structure it lives in.
- Wheel condition on each shuttle: flat spots, chunking, polyurethane delamination, debris embedded in the tread. Wheels are the highest-wear item on the machine and the one that goes from "looks fine" to "positioning drift on every cycle" fastest.
- Rail surfaces and lane-end stops in the lanes that ran the highest cycle counts that week — scoring, build-up of wheel material, loose fixings. Deep-lane systems concentrate wear at the entrance of every lane, not evenly along it.
- Sensor and lens cleaning on the shuttles, with particular attention to any that cross a temperature boundary. Condensation on an optical sensor reads as a false stop, and the fault code will send you looking at the wrong thing.
- Charging contacts and connectors — pitting, carbon build-up, bent pins. A charging fault presents as a battery fault and gets a battery replaced for nothing.
Monthly
The monthly round is where measurement starts to replace inspection: you are not asking "does it look all right", you are asking "what is the number, and which way is it moving compared with last month".
- Battery health per shuttle — capacity against nameplate, charge-cycle count, internal resistance where the BMS reports it. Plot it. A battery that has lost capacity steadily for three months is a planned replacement; the same battery discovered dead is an emergency and a stranded pallet.
- Drive current draw and motor temperature at a reference load on each shuttle, compared with the fleet and with that shuttle's own history. A drive pulling more current to do the same work is telling you about a bearing, a brake dragging, or a gearbox — before it tells you by stopping.
- Lift and transfer car: hoist rope or chain condition, guide roller wear, level and alignment at each rack level, brake function under load. On most pallet shuttle systems the lift is the true single point of failure, and it is also the piece that falls under lifting-equipment regulation.
- Structural fixings on the rails and lane-entry guides, and a torque check on a sample of them. Thermal cycling in cold storage works fasteners loose slowly enough that nobody sees it happen.
Quarterly and annually
The longer intervals carry the manufacturer's own service schedule — gearbox lubrication, drive-belt replacement, bearing changes at a stated cycle count — and the regulatory items. In the UK, a shuttle lift or crane is lifting equipment under LOLER and needs a thorough examination by a competent person at the statutory interval; the shuttles, the lanes and the racking fall under PUWER. The same logic applies under the equivalent regimes elsewhere in Europe and North America: the lifting element has an external examination cadence that does not care how busy the site is.
- Quarterly: drive-component inspection per shuttle (bearings, belts, couplings), full rail-alignment survey on the highest-cycle lanes, firmware and controller fault-history export for trend analysis.
- Annually, or at the OEM's stated cycle count, whichever comes first: wheel replacement on high-throughput lanes, battery replacement where capacity has crossed the manufacturer's threshold, and the full manufacturer service on the lift.
- Statutory: the lifting-equipment thorough examination, documented, with the report filed against the asset rather than in an inbox.
Why the checklist usually is not the problem
Most shuttle sites already have something like the list above. What they rarely have is the version that matches the manual. The OEM documentation for a shuttle system runs to hundreds of pages across the shuttle, the lift, the WCS and the racking, and the service intervals are scattered through it — a torque figure in one chapter, a cycle-count limit in an appendix, a battery threshold in the BMS supplement. Somebody transcribes what they can find in the week before go-live, and that becomes the programme.
STIQ's 2026 goods-to-person research tracks the market moving toward denser, more shuttle-dependent storage. More sites are inheriting more of that documentation, and the gap between the checklist on the wall and the one the manufacturer actually specified is where the warranty disputes and the stranded lanes come from.
How we approach it
Reliabilytics starts from the manual rather than from memory. Upload the OEM documentation for the shuttle, the lift and the racking, and the AI extracts every service interval, procedure, torque value and safety-critical check into a PM schedule linked to each asset — so the quarterly gearbox service on page 148 exists as a work order with a due date, not as a line someone meant to add.
Condition monitoring then keeps the schedule honest between inspections: battery capacity, drive current and positioning-correction frequency, from dedicated sensors or from the telemetry the shuttles already produce, with a work order raised when a trend crosses the threshold rather than when a lane stops. And every completed check is timestamped with who did it, what they found and what parts went in — the record that answers an OEM warranty assessor, a LOLER inspector, or the question of why lane 14 keeps faulting.
None of this changes what needs checking. It changes whether the checklist you are running is the one the manufacturer wrote, and whether you can prove it was done.