AMR Spare Parts Planning: What to Stock Before a Critical Robot Fails

The spare is on the shelf. The repair is still waiting.

At the start of a night shift, a mobile robot is removed from service after a drive-related fault. Maintenance finds a replacement assembly in the storeroom. The carton has the expected product-family name, but its hardware revision differs from the installed unit. Nobody on duty can establish whether the replacement is approved for that robot configuration. The supplier's specialist will be available the following morning.

This fictional incident exposes the purchasing problem behind AMR spare parts planning. Physical possession is only one condition of repair readiness. The replacement must be suitable, available for withdrawal, installable by the authorized team, and supported by the configuration information and checks required to return the machine to service.

A factory can spend heavily on a parts cabinet while retaining the same practical dependency on a distant specialist. It can also buy an extensive service package while leaving essential hardware outside the provider's available stock. The useful purchasing objective is a complete, executable support arrangement for the failures that matter to production.

This guide follows that objective through stocking decisions, a repairable-parts example and the commercial details that determine whether support works after commissioning. It concerns parts used to maintain autonomous mobile robots and their supplied equipment. It does not concern robots delivering spare parts for unrelated machinery.

The operating response to a stopped robot remains a separate engineering task. Our guide to AMR service recovery covers material-state preservation and restoration of transport service. Here, the decision is where to hold replacement resources, how to replenish them, and which obligations belong in the purchase.

Give each stocked item an operational reason to exist

Start with the installed equipment register and the approved maintenance documentation. Include the robot base, top module, charging equipment and other supplied assets that the maintenance team is expected to support. Record the service owner for each. A critical lift assembly supplied by an integrator can be overlooked when the buyer requests only the mobile-base manufacturer's standard spare kit.

Next, identify the replaceable assembly that the local team is actually permitted and equipped to change. The component named in an alarm may sit inside a larger assembly that requires workshop repair. Buying the smallest apparent failed part can leave the plant without the assembly, tools or procedure needed for an approved field replacement.

For each candidate, ask which mission loses capacity, how long the operation can tolerate that loss, and what validated alternative remains available. A high-priced part supporting a deferrable route may justify regional stock. A modestly priced part unique to a critical load interface may justify local stock even when failures are infrequent.

The priority of critical spare parts should therefore follow consequence and usable replacement lead time. Purchase price and historical consumption remain relevant, but neither describes production exposure alone. Low consumption may reflect a reliable design, a small installed population or a short observation period. Establish which explanation the available evidence supports.

Separate scheduled consumption from uncertain withdrawals

A useful AMR maintenance plan distinguishes work already scheduled from corrective demand that may occur unexpectedly. A planned replacement consumes a known quantity on a known date. Reserve that quantity explicitly and arrange its replenishment. Otherwise, the workshop can consume the emergency stock during routine service while the inventory dashboard still implies that contingency coverage exists.

Use manufacturer instructions and the actual duty profile to schedule maintenance. This article does not prescribe a universal wheel, bearing or battery replacement interval. Relevant evidence can include inspection findings, operating hours, distance, load conditions and the applicable maintenance schedule. Similar-looking robots used on different surfaces may require different planning assumptions.

Keep a distinction between routine wear items, repairable assemblies and rarely used contingency items. Their stocking logic differs. Consumables may be ordered around scheduled demand; repairable units circulate through a return process; contingency stock is held because an unacceptable waiting time would otherwise remain. Combining all three into one annual parts allowance hides those decisions.

Choose the storage location together with the service route

An AMR spare parts inventory can be distributed across several locations. The appropriate arrangement depends on how quickly a usable unit can reach the job and what happens after that unit is withdrawn. Ownership, physical location and entitlement to draw stock are separate questions. A supplier may own an item stored at the plant; the plant may own an item held at a regional warehouse.

Local stock creates access and a preservation obligation

Plant-owned stock gives the maintenance team direct access when withdrawal procedures, approvals and shift coverage permit it. The buyer also accepts storage cost, inventory accuracy and obsolescence exposure. Verify that the storeroom can maintain the supplier's specified conditions and perform any required periodic checks. A sealed carton with an unreadable status label should not automatically count as a serviceable spare.

A relevant AGV example appears in AGVE's published service offering, which describes longer service contracts using spare stock at the customer site. The public description supports the existence of that service approach, but leaves the stock quantity, ownership and replenishment commitment to be established. Those details are essential inputs to the buyer's coverage decision.

Consignment can change who owns the item before use, but it still needs practical rules. Establish who may release it, when consumption becomes billable, who inspects it and how quickly the supplier replenishes it. Confirm whether the item is allocated exclusively to the site or can be recalled. The financial arrangement does not by itself establish availability during a night-shift incident.

Regional stock needs an allocation and transport commitment

A regional depot can pool expensive assemblies across several installations. Ask where the stock is held, whether the relevant revision is included, and how competing requests are prioritized. An inventory total across a global network provides little guidance about what can reach one plant before its approved fallback capacity is exhausted.

Break delivery into the steps the supplier actually controls: order acceptance, technical identification, allocation, warehouse release, collection and transport. Confirm cutoff times, weekends and destination receiving hours. A promise to dispatch within a stated interval should be recorded as dispatch performance. The project needs a separate understanding of arrival and readiness for use.

Advance exchange and repair return create different waiting periods

Under an agreed advance-exchange arrangement, a serviceable unit is supplied before the failed unit has completed repair. Confirm the eligibility criteria, stock commitment, core-return deadline and charges for an unacceptable or missing return. Ordinary repair return may require the plant to send the failed item away and wait for assessment, repair and shipment of that same unit.

For AMR replacement parts, ask suppliers to identify the offered route item by item. A general statement that repairs are available does not establish access to an exchange pool. Some items may have different arrangements because of configuration, repair capability or commercial scope. Put those exceptions beside the affected part numbers.

For example, AGVE's published repair-request document describes an ordinary repair route that returns the submitted unit. Its separate limited-warranty subsection allows repair or exchange where possible. The document also requests part and serial identity, observed symptoms and requested programming. Read the applicable arrangement carefully before assuming that a return authorization releases another unit for immediate shipment.

A spare robot covers a different set of dependencies

A complete reserve vehicle can reduce the need for immediate field repair when it is qualified for the required missions and can be introduced through the approved recovery process. It still requires charging, maintained configuration, compatible attachments and periodic readiness checks. It may not replace a unique top module or resolve a failed shared station.

Compare a reserve vehicle with a parts pool only over the failures each option actually covers. The vehicle may address several independent hardware failures but leave common software, infrastructure or attachment problems untouched. Avoid assigning the same avoided downtime to several overlapping investments when calculating their combined value.

Follow the pool through the second failure

Consider a fictional factory with eight robots that use an approved common drive assembly. The factory has one serviceable spare, and its trained maintenance team can perform the documented replacement and release procedure. Assume the transport system can tolerate one unavailable robot temporarily. These assumptions define this example; they need evidence in a real installation.

The supplier repairs the removed unit and returns it to stock. The planning estimate is ten calendar days from withdrawal to receipt of a serviceable, accepted replacement. That interval includes internal handling, shipment, supplier work, return transport and receiving checks. It describes replenishment of the spare pool, not the hands-on time needed to replace the assembly.

At the beginning of Day 0, the first failure consumes the spare. The affected robot returns to service after its approved replacement process, and the failed assembly enters the repair pipeline. The local serviceable balance is now zero. On Day 4, a second robot requires the same assembly. The first repaired unit is not due back until Day 10.

For this example, the second failed assembly remains in the isolated robot until the replacement arrives on Day 10. Its own repair-return process starts after removal. This assumption keeps the two physical units and their replenishment records distinct; a plant that removes and ships the second assembly earlier should model that different sequence.

The second requirement waits six days for a part in this constructed sequence. That is a stock-related waiting interval, not a prediction of six days of stopped production. Actual production consequences depend on remaining fleet capacity, buffers and approved alternatives. Replacement and release time must also be considered before the second robot becomes available.

The important observation occurs on Day 10. Receiving the repaired unit closes the first supplier return, but immediate issue against the waiting requirement leaves the plant with zero ready stock again. A purchasing report showing that the repair order is complete can coexist with an unresolved operational exposure.

Change one assumption and compare the outcome

Two initial serviceable spares would satisfy both withdrawals in this exact sequence. That result does not establish that two is the correct stock level for the fleet. A third overlapping requirement, an incompatible return or a repair rejected as uneconomic would produce a different outcome. Stock quantities should be tested against plausible sequences, not inferred from one favorable example.

Alternatively, suppose the first withdrawal triggers a guaranteed compatible replenishment accepted on Day 2. One initial spare then covers the Day 0 and Day 4 requirements in this sequence. The benefit comes from the shorter replenishment exposure. It depends on the supplier delivering the correct serviceable item and the plant completing its receiving process within the assumed interval.

A third option is faster diagnosis or replacement labor. That can shorten the robot's initial outage while leaving the ten-day pool-replenishment interval unchanged. Separate the improvement being purchased. Faster field work, faster replenishment and a larger serviceable pool solve related but different constraints.

Use demand evidence without pretending that an average is a guarantee

Suppose the same example fleet is assumed to require six corrective assembly withdrawals per year, evenly represented by a simple average. Over a ten-day replenishment interval, average demand is 6 × 10 ÷ 365, approximately 0.164 units. Rounding that result up to one does not establish a service level or rule out two closely spaced withdrawals.

This arithmetic describes average exposure only. A stockout probability would require an appropriate demand model and its assumptions. A new fleet may lack enough observations to support one. Begin with transparent scenarios and revise them using verified withdrawal, diagnosis and replenishment records rather than presenting uncertain inputs as precise reliability data.

Check for clustered causes. Similar robots may encounter the same floor damage, contamination, maintenance error or faulty batch. Planned campaigns can also create concentrated demand. A pooled spare arrangement must consider these shared exposures; treating every withdrawal as an unrelated event can make the proposed coverage look stronger than it is.

Price the extra coverage with a break-even question

Compare support options against a stated baseline. For a second illustrative calculation, assume a faster replenishment arrangement costs an additional $4,800 annually. Finance accepts $600 per hour as the incremental consequence of an uncovered parts delay for the defined critical operation. Both figures are invented inputs, and the cost rate applies only while that specific production consequence exists.

The arrangement would need to avoid $4,800 ÷ $600 = eight equivalent hours of that consequence per year to break even on this narrow measure. This is a decision threshold, not an assertion that the service will prevent eight hours. The purchasing team must establish which plausible incidents the arrangement changes and how much consequence remains after approved fallback measures.

If the effective consequence were $200 per hour, the threshold would rise to 24 hours. If an already-funded reserve vehicle preserves all required production during the parts delay, the incremental production benefit could be much smaller. There may still be value in reducing overtime, replenishing resilience sooner or avoiding expedited freight, but those benefits need separate evidence.

For a locally owned assembly, include the initial cash commitment, storage and inspection work, expected repair or replacement expenditure, and the risk that the item becomes unusable or obsolete. Do not automatically treat the entire purchase price as a recurring annual expense. If finance uses an annual carrying-cost rate, state its components and avoid counting the same financing or obsolescence allowance twice.

Place these assumptions inside the wider AMR lifecycle cost model. Our guide to AMR maintenance and ownership costs provides the broader investment boundary. The spare-parts comparison should explain the incremental difference between support options rather than rebuild the whole robot business case.

Keep committed cash expenditure, expected repair demand and uncertain interruption consequences visible as separate lines. This allows purchasing to negotiate the service price while operations tests the consequence assumptions. A single blended cost per robot conceals which input actually determines the decision.

Also ask what each option cannot address. Additional stock may offer little benefit if diagnosis requires an unavailable specialist. A shorter depot lead time may not help when the item needs configuration at a separate workshop. Fund the limiting resource identified by the scenario before expanding a cabinet simply because its contents are easy to count.

Translate the support package into events that can be measured

An AMR service agreement should connect covered assets, service hours, obligations and evidence. Define incident severity through operational impact and approved alternatives. One stopped vehicle in a lightly loaded fleet may have a different consequence from failure of the only qualified vehicle for a critical mission. Give both parties a clear method for classifying and escalating the incident.

Current public offerings show why package details matter. MiR's service page distinguishes the standard warranty from additional MiR Care coverage and lists technical support, parts and repairs among its service offerings. That describes available service categories. The buyer still needs the applicable quotation and terms to establish its own coverage, exclusions and delivery commitments.

Similarly, OTTO Care's public page describes weekday support for Plus and around-the-clock support for Pro, with advertised response times as fast as four hours and 30 minutes respectively. These are response descriptions, not promises that a failed robot will be repaired within those intervals. Confirm the relevant contractual target and conditions for the proposed site.

These are proposed measurement boundaries, not universal service definitions. Agree which events are contractual commitments and which are monitoring points. Diagnostic duration may depend on information that is unavailable at first contact. Recording that uncertainty is more useful than assigning an optimistic resolution target without a workable evidence path.

For robot repair turnaround time, specify whether the interval starts when the plant removes the assembly, when the supplier receives it, or when a repair quotation is approved. Likewise, distinguish shipment back from acceptance into serviceable stock. A five-day workshop target can sit inside a much longer total replenishment interval.

Make the calendar and the responsibilities explicit

State the time zone, working calendar, holiday coverage and escalation route. If a clock can pause while awaiting customer information or access, record the reason and the start and end of the pause. Retain the full elapsed incident time as well. This allows the plant to see its actual exposure even when the provider has met a narrower contractual measure.

Ask who supplies labor, travel, freight, consumables, software access and post-replacement checks. Confirm treatment of attachments and equipment supplied by other parties. A service package covering the base robot may leave the production-critical load mechanism under a separate agreement. One incident coordinator should be able to connect those responsibilities without waiting for a commercial dispute to be settled.

Link the response schedule to the existing AMR RFQ scope. Include the initial parts list, training, replenishment arrangements and named service boundaries as priced deliverables. The buyer should be able to trace a support commitment from proposal to order and then to the maintenance team's operating documents.

Service credits, where agreed, compensate under defined commercial terms; they do not supply the missing assembly or release a repaired robot. Judge the operational arrangement by the resources it makes available. Use contractual remedies to support accountability while retaining a practical plan for the incident itself.

Run the spare pool as part of daily maintenance

Effective mobile robot maintenance needs a current view of both installed assets and serviceable stock. Assign distinct statuses for available, reserved, awaiting inspection, quarantined, in transit, under repair and retired items. A failed assembly sitting in the return area remains physical inventory, but it cannot satisfy an immediate replacement requirement.

Link withdrawals to an asset and a work order. Record the symptom, diagnostic conclusion, removed serial number, installed serial number and disposition of the removed unit where applicable. WAKU Robotics' spare-parts management offering explicitly describes links to work orders, visibility across locations and records of repair and calibration activities. Those are relevant software capabilities; their effectiveness still depends on accurate local records and execution.

A received part must become a usable part

Check the received identity, revision, condition and required documentation against the approved application. For configuration-dependent assemblies, establish what must be loaded, transferred or verified before installation. Do not assume that a connector match or a broadly similar model name establishes interchangeability. Obtain supplier confirmation for the actual installed configuration.

The release process should follow the applicable manufacturer and integrator instructions. Some replacements require functional, calibration or safety-related verification beyond the physical fitting operation. Allocate the authorized personnel, tools and access required for that work. The site's sensor replacement and calibration checks explain one important part of that boundary.

A model-specific example is documented in the OMRON LD Platform OEM User's Guide, 11970-000 Rev K, page 138: replacing the core requires the stated emergency-stop and safety-laser commissioning procedures to be repeated. This older manual illustrates a replacement dependency; it is not presented as the current procedure for every OMRON robot. Obtain the applicable instructions for the actual model and revision.

When a supplier supersedes a part, record whether the replacement is directly applicable, needs a conversion kit, requires software changes or applies only to certain vehicles. Connect the decision to AMR configuration change control. Update the stock register and purchasing instructions so the next urgent order does not recreate the same uncertainty.

Replenishment starts when stock is withdrawn

Name the person or system that initiates replenishment, obtains the return authorization, prepares the shipment and tracks the supplier's response. Delays before shipment belong in the replenishment record. A technically efficient repair service cannot compensate for a failed assembly left unprocessed in a local cage for several days.

Define what happens when repair is rejected, delayed or declared uneconomic. The contingency may be a new purchase, an agreed exchange or temporary coverage from another location. Establish approval authority and an escalation date before the pool reaches zero. Each additional approval step should have an owner who is available during the relevant operating calendar.

A no-fault-found return also needs review. The reported symptom may have involved a cable, an intermittent condition or an interaction outside the returned assembly. Preserve the diagnostic evidence and confirm the appropriate next action before placing the same item back into unrestricted available stock. Repeated swaps without a verified cause can consume the pool while leaving the underlying problem active.

Test access before the first urgent withdrawal

During handover, conduct a controlled support rehearsal. Ask the actual shift team to identify a candidate replacement, locate its eligibility record, access the required instructions and explain how replenishment would begin. Exercise the agreed support channel at an authorized time. A document review or planned maintenance task can reveal access gaps without deliberately damaging equipment.

Use the rehearsal to confirm that AMR service support is available to the people who will need it. Check account permissions, contact coverage, purchasing authority and receiving arrangements. If remote assistance requires approved connectivity, establish that access through the plant's normal controls rather than discovering the dependency during an outage.

Review the arrangement after meaningful changes: fleet expansion, a new duty cycle, a major component revision or a changed service provider. Look at serviceable fill performance, actual replenishment intervals, emergency purchases and repeated unresolved faults. With limited incident counts, show individual cases rather than presenting unstable averages as dependable forecasts.

The maintenance handover should leave the plant with a usable pool, a replenishment owner and a support route that has been exercised. A parts cabinet and a telephone number are inputs. Their value emerges when the next qualified replacement can be completed and the depleted stock position can be restored.

Focused FAQ

Should every robot have its own full spare-parts kit?

No universal ratio is appropriate. Shared stock may cover several robots when the parts are approved for their configurations and replenishment exposure is acceptable. Dedicated stock may be justified for a unique attachment or an isolated location. Make the decision by replaceable assembly, operating consequence and available alternatives, then test overlapping withdrawal scenarios.

Does a warranty remove the need to hold local stock?

A warranty can define responsibility for specified defects and remedies. The applicable terms determine what is covered. It does not automatically establish that a compatible part, authorized technician and release capability will be available within the plant's tolerated interruption. Review that operational requirement separately from who ultimately pays for the repair.

Can two factories share one expensive spare?

They can evaluate that arrangement if compatibility, allocation priority, transport and receiving conditions are documented. Both sites must understand what happens during simultaneous demand. Count the transfer time and any configuration work, and confirm that removing the item from the shared pool triggers replenishment. Shared ownership alone does not create serviceable coverage at both locations.

How should a new fleet be planned without failure history?

Start with the manufacturer's recommendations, applicable installed-base evidence and explicit operating scenarios. Record uncertainty in withdrawals, repairability and replenishment time. Use a limited initial arrangement that can be revised as reliable observations accumulate. Avoid converting a small number of incidents, or a vendor's broad reliability statement, into an unsupported statistical service guarantee.

What should be reviewed when a part approaches end of supply?

Ask for the affected configurations, available replacement route, timing and support implications. Compare a final stock purchase with conversion or planned equipment replacement. Include storage life, required testing and the remaining operating horizon. A large final purchase can leave the plant holding unusable stock if software or configuration support ends first.

Evidence scope: linked primary sources substantiate only the attributed service offerings and documented examples. The stocking method, fictional incident, numerical scenarios and purchasing analysis are original editorial work. Sources were reviewed on September 24, 2026; actual coverage, replacement procedures and commitments must be established in the applicable supplier documents.

#AMRSpareParts #RobotMaintenance #SparePartsPlanning #AMRService #MaintenanceProcurement #Intralogistics #IndustrialAutomation #FleetReliability