Europe can retain more of the work already invested in products by improving fault diagnosis, recovering usable components and qualifying them for another service life.
The value inside a worn component
A failed machine contains more than scrap. Its castings have been formed, surfaces machined, parts assembled and controls integrated. Recovering the metal preserves one layer of that investment. Returning a suitable component to service can preserve several. The difficult decision is identifying how much useful function remains and what must be replaced before a customer can rely on it.
Repair addresses a fault. Remanufacturing involves a more extensive industrial process, commonly including disassembly, inspection, restoration and testing against a defined specification. Neither route makes every used product suitable for another life. Their commercial strength depends on selecting the right intervention before damage, obsolescence or the cost of investigation consumes the value being saved.
The scale is already industrial in some sectors. Renault reports that THE REMAKERS gave 350,000 parts, drawn from 11,000 references, a further life in 2025. Its Flins site also retrofitted more than 150 production robots that year. These are company operating figures, rather than estimates of everything technically recoverable from Europe’s installed machinery.
Diagnosis changes the repair decision
A product’s age alone is a poor guide to the intervention it needs. A Joint Research Centre study published in 2019 analysed around 11,000 diagnoses of washing machines and dishwashers from an independent European repair centre. Washing machine doors and carbon brushes had comparatively high repair rates; bearings, drums and tubs, circulation pumps and electronics had lower rates. The dataset describes machines brought to that operator, not a representative failure census of every appliance sold.
That distinction has practical value. Counting how many products could theoretically be opened misses the combination of failure, part price and labour that determines whether a repair proceeds. Manufacturers need feedback on rejected repairs as well as successful ones. A bearing assembly that repeatedly makes otherwise functional machines uneconomic to repair is a different design problem from a readily replaceable door component.
For service businesses, the useful record therefore starts before a replacement part is ordered: symptoms, diagnostic steps, confirmed fault, quoted cost and the customer’s decision. Comparing those records can reveal whether the constraint is inaccessible hardware, missing information, an expensive module or a service network that cannot deliver quickly enough.
Designing access to the parts that matter
The Joint Research Centre’s repair scoring research treats disassembly depth as the number of steps needed to remove a part and stresses reversible access, allowing reassembly without damage to functional components. That makes repairability more specific than a claim that a product can be dismantled. Breaking a housing to recover its metal is fundamentally different from opening it to replace a failed component.
The engineering priority should follow likely failures and the consequences of reaching them. A serviceable pump behind a removable cover may be more useful than a product with many detachable parts but a sealed critical assembly. Standard tools, identifiable fasteners and a clear sequence reduce uncertainty for technicians, although the final choice must still account for sealing, vibration and other operating requirements.
Access also has a commercial dimension. A repair manual without an available replacement is incomplete; a stocked component without a reliable diagnostic procedure can lead to unnecessary swaps. Product teams and service operations therefore need a shared view of the repair, including the time required to identify the fault and verify the completed work.
Factories need cores as well as customers
Remanufacturing adds another supply chain: the return of used products or components, often called cores. Caterpillar’s core exchange model links credit towards a replacement to acceptance criteria. A core requiring more salvage work is less valuable than one meeting the relevant conditions. The returned object is consequently an input with a variable quality and price, rather than an interchangeable quantity of waste.
This changes production planning. A factory can receive enough units by weight yet lack the particular housings, assemblies or versions needed to fulfil orders. Return timing, identification and protection during transport become part of manufacturing performance. A damaged connector or corrosion acquired in storage may destroy an opportunity that existed when the component was removed.
Purchasing and maintenance teams can help by agreeing how cores will be handled before a breakdown occurs. Exchange stocks can shorten equipment downtime, but someone must finance those stocks and bear the risk of obsolete variants. A viable service balances the value recovered with transport, inspection, replacement parts, testing and the inventory needed to make the offer dependable.
A useful factory dashboard would separate units received, accepted, restored and sold, with reasons for rejection at each stage. Combining these into a single recovery rate obscures whether a business is improving its process or merely receiving easier inputs. It also helps distinguish a shortage of customer demand from a shortage of repairable cores, which require quite different responses.
Qualification is the product
ABB’s published robot remanufacturing process begins with traceability checks, followed by disassembly and a 152-point inspection covering robots and controllers. Components that fail inspection are replaced, and assembled robots undergo at least 16 hours of testing under nominal load. These are the manufacturer’s stated procedures; they do not establish that any used robot is suitable for any new application.
Bearings illustrate why qualification also depends on the intended duty. SKF describes checking damage depth, clearance, ovality and variation in ring wall thickness before selecting remanufacturing work. Additional nondestructive testing can depend on the specification or customer request. The company emphasises matching the procedure to the application. SKF bearing assessment
The buyer is purchasing a documented capability, not simply an object that looks restored. Inspection records, tolerances and acceptance tests should travel with the component. For a robot moved to a different task, integration still needs attention to payload, reach, controls and the complete cell. For a bearing, the relevant question is whether its assessed condition fits the required load and service interval.
Automation must match the returned product
Some repetitive service operations are becoming more automated. In September 2026, Michelin introduced NAZCA, an automated regrooving system for suitable truck, bus and coach tyres. Its evidence included more than 400 operations during proofs of concept across three Euromaster branches, with a cumulative year of testing. The company scoped its hands-on time estimates to normal operating conditions without issues.
Regrooving extends use of an appropriate existing tread; retreading subsequently adds a new tread to a qualified casing. Neither means that any worn tyre can be processed indefinitely. The example shows where automation can assist: a defined intervention, compatible inputs and repeatable operating conditions. Mixed, damaged or unidentified returns demand a broader range of decisions.
Investment cases should therefore include exceptions. A fast automated station may still require people to assess unsuitable inputs, change tooling or handle incomplete records. Measuring output across the whole workshop, including inspection and rejected items, gives a more useful picture than the cycle time of the successful operation alone.
Measure the service that survives
Environmental comparisons require similarly clear boundaries. In September 2026, The Future is NEUTRAL reported Quantis assessments of eight families of remanufactured automotive parts. For the studied electric motors, reported reductions were at least 67% across five environmental indicators relative to new counterparts. The company said the assessments underwent external critical review and used a cut-off approach: remanufacturing and transport were counted, while original production burdens were not allocated to the returned part.
Those findings support a defined comparison, not a universal percentage for repaired equipment. Buyers should ask what service life is assumed, which operations are included and whether the replacement performs the same task. Preserving an assembly is valuable only if the resulting product delivers useful service.
For European industry, the opportunity is to organise around that service: design critical parts for access, recover suitable cores promptly and sell a qualified second life. The most revealing measure is not how many tonnes entered a workshop, but how much dependable function left it.
