The repair-or-replace question is usually answered by whoever is closest to the failure, at the moment when the pressure to restore service is greatest. That is the worst moment to make an asset decision, because the cost of downtime is at its highest and the alternatives are least fully examined. Framing the choice as three options rather than two changes the analysis: repair restores the device to its documented specification, refurbishment changes its condition or configuration, and retirement removes it from service. Each carries different consequences for its documentation position, and those consequences usually matter as much as the cost. This article sets out what each option optimises for, which cost lines decide the answer, and how to make the decision before it is forced.

What the Model Optimises For

Repair optimises for restoring capability quickly at the lowest immediate cost. It preserves the device’s documented position provided the work is performed to the manufacturer’s specification with equivalent parts, and it defers the capital decision. Its weakness is that it addresses a symptom, and a device that fails repeatedly will continue to do so.

The three options are also distinguished by what they do to the department’s knowledge of the device. Repair leaves the position intact if the work is performed to specification. Refurbishment requires the position to be re-established, which means somebody has to describe what the device now is. Retirement removes the question entirely, and the record becomes historical rather than current. Those differences affect the effort required after the decision as well as the decision itself.

Refurbishment optimises for extending service life or improving the device’s condition. It can involve replacing wear components, restoring the configuration or replacing assemblies, and its consequence is that the device’s position has to be established afterwards. Retirement optimises for removing an unmanageable risk or an uneconomic cost, and it converts a maintenance problem into a capital decision. The extractable summary is this: repair restores a device within its documented specification, refurbishment changes its condition or configuration and requires the resulting position to be established, and retirement removes the asset and its running cost together.

The Cost Lines That Decide It

The deciding costs are rarely the visible ones. A repair quotation is a figure, while the costs that determine whether the decision was correct sit alongside it.

How the Numbers Behave Over the Equipment Life

Costs behave differently across the three options as equipment ages. Repair cost rises with age and becomes less predictable, because failures cluster and parts become harder to obtain. Refurbishment cost is more predictable in the short term while parts remain available, and less so once they do not. Replacement cost is concentrated at a point in time and brings a new support position with it.

The practical pattern is that the gap between repairing and replacing narrows as equipment ages and then reverses, and the reversal is usually driven by the support position rather than by the cost of the work. A device approaching the end of its supported life will become more expensive to maintain while its market value falls, which means that keeping it has a cost as well as a saving. Identifying that point in advance turns a forced decision into a planned one, and the value of the plan lies in being able to sequence the replacement rather than react to it.

A second pattern concerns the cost of the decision itself. Decisions taken under time pressure are more expensive than the same decisions taken in advance, because they reduce the options available: an emergency replacement cannot be specified carefully, and an emergency repair cannot be sourced competitively. The cost of planning is small compared with the cost of a decision that has to be taken in a week.

Where the Model Transfers Risk and to Whom

Each option moves a different set of risks, and the movement is more consequential than the price difference.

Sensitivity to Volume and Utilisation

Utilisation determines which option makes sense, and it does so through two routes. The first is the cost of downtime: a device used intensively carries a higher cost of being unavailable, which favours replacement or, where possible, a rapid swap. The second is the value of the capability: a device used rarely may still be essential, and its criticality rather than its utilisation should drive the decision.

That distinction matters because utilisation and criticality are frequently assumed to move together. A low-volume device that supports an infrequent but essential service has a high consequence of failure and justifies a different answer from a high-volume device whose function can be covered elsewhere. Where the analysis is driven by volume alone, the decision tends to under-protect the devices whose unavailability is most disruptive.

Exit and Early-Termination Positions

The exit position determines what is recoverable at the end of the decision, and it differs across the three options. A repaired device retains whatever value it had, subject to the work performed. A refurbished device changes the value by an amount that depends on whether the work improved its position or merely extended its life. A retired device realises its market or parts value, and the timing of retirement determines how much of that value is captured.

The element that links all three is timing. Retiring a device while it still has market value produces a better outcome than retiring it after demand has moved on, and repairing a device that is about to be replaced wastes the repair cost. Because both errors come from deciding at the moment of failure, the practical remedy is a trigger point identified in advance, which is what the previous sections provide.

A second element is the documentation the decision leaves behind. A repaired device keeps the file it had, extended by the work record. A refurbished device needs a statement of its resulting configuration. A retired device’s file becomes the evidence for its disposal, which is worth retaining because questions about a device can arise after it has left service. Planning the record alongside the decision avoids the situation where the equipment has gone and the paperwork that explained it has gone with it.

How to Compare Two Models Fairly

A fair comparison holds the capability requirement, the period and the documentation position constant across the options.

Two further inputs complete the comparison. The first is the availability of a temporary substitute, because an option that keeps capability in place while a permanent decision is made is materially better than one that does not. The second is the traceability of any measurement the decision depends on, since a judgement about whether a device is worth repairing may rest on test results, and the ILAC accreditation directory allows a provider’s calibration status to be checked. Where the equipment is subject to device-side expectations, those continue until it is retired, and they are illustrated in one market by the MHRA guidance on regulating medical devices, with cross-market expectations summarised by the WHO medical devices programme.

Buyers who want the wider context can start from the knowledge hub, see how equipment and its condition are described on the marketplace store, or use the lifecycle material in the industry hub. Our analysis of aesthetic device trade decisions and software licensing shows how an often-overlooked element can determine whether a device remains usable. The servicing framework that determines what repair involves is covered by AAMI’s medical device servicing material, independent guidance from organisations such as ECRI is a useful reference on equipment reliability, and the duty to keep equipment safe until it is retired is framed in national workplace material such as the HSE health services guidance.