Spare parts management: deciding what to hold before the machine needs it

Stock the consequence, not the catalogue

The case for holding a part rests on what happens if it is not there: how long the equipment would be down, whether production can be moved elsewhere, and how long the supplier takes to deliver. A cheap component with a long lead time on a machine that limits the plant is worth holding; an expensive one for a machine with a standby is often not. Work through the critical assets and their plausible failures rather than through supplier recommended lists, which are written to be comprehensive and tend to include everything the equipment contains.

Insurance spares are a separate category with a separate decision

Some components fail rarely, cost a great deal and take months to obtain: a large motor, a gearbox, a custom fabricated assembly. Holding one is an insurance decision made at a senior level, weighed against the production loss during an outage and against alternatives such as a shared arrangement with another site, a supplier holding agreement, or accepting the risk. What should not happen is this decision being made by default, either by a storekeeper who cannot justify the value or by an engineer ordering a spare that then depreciates unnoticed for a decade.

Storeroom control and the private stashes that defeat it

Parts issued without record, taken from the store out of hours, or kept in a technician's cupboard make the recorded stock useless, and the recorded stock is what the next person relies on during a breakdown. The remedy is a system people can actually use at the moment they need a part, including out of hours, plus periodic counting of the critical items. Personal stores usually indicate that the official arrangement has failed someone before. Ask why before instructing people to hand parts back, since the underlying availability problem will simply recreate them.

Obsolescence creeps up quietly on long-lived assets

Machines outlive their control systems, their drives and their manufacturers. Parts become unavailable, suppliers exit, and equipment reaches a point where a common failure cannot be repaired at any price. Review the older critical assets specifically for this, ask suppliers about support status, and decide deliberately whether to hold a last-time purchase of key components, arrange a rebuild capability, or plan replacement. The corresponding discipline is disposal: a periodic review to write off parts for equipment no longer on site, which is uncomfortable but cheaper than the space and the counting effort they consume.

The parts record has to work during a breakdown at night

Spares data fails in predictable ways: parts identified only by a supplier code nobody recognises, no link between the part and the machines it fits, duplicate records for the same item under different descriptions, and a location that was accurate before the store was rearranged. The test is whether a technician who does not normally work on that machine can identify and find the part quickly at an inconvenient hour. Building the machine-to-part links takes deliberate effort during normal working, and it is the difference between a short stoppage and a long one.

Frequently asked questions

Should spare parts sit on the maintenance budget or as stock?

Treat them as stock with a clear owner, because expensing parts on purchase encourages buying at the point of failure and discourages holding anything. Whichever accounting treatment applies, someone must be accountable for the value held and for reviewing it, and that person needs enough engineering input to distinguish a part that protects the plant from one that simply accumulated. Finance and engineering disagreeing about this is common and worth resolving explicitly.

Can we rely on suppliers holding spares for us?

Only where the commitment is written, specific about which parts and what response time, and periodically verified. Verbal assurances of availability tend to evaporate when several customers need the same item. Where a supplier arrangement is genuinely reliable it removes real cost, particularly for expensive slow-moving components, but the risk shifts to their stock policy and their continued existence, both of which deserve the same scrutiny as any other single point of dependency.

How do we start improving spares control in a plant with poor records?

Do not attempt to catalogue everything first. Begin with the assets that limit output, list their plausible failures, and check whether the parts those failures need are on site and findable. That produces a short, high-value list and some immediate gaps worth closing. Broaden gradually from there, and use each breakdown as an opportunity to correct the record for the parts involved, which builds accurate data where it is actually used.