This spare part has not moved in years. Should you really remove it?

The inventory report has found an expensive component with no recorded issue for several years. An assistant puts it on the disposal list. The warehouse would recover space and finance would reduce obsolete stock.

Before agreeing, the maintenance engineer asks which machine still depends on it.

That question changes the analysis. The component may belong to equipment already retired. It may also be the only replacement for a running machine whose supplier no longer manufactures the part.

Low consumption tells you how often the store has issued an item. It does not, by itself, tell you why the item is there.

Separate the decisions hiding inside one list

A slow-moving-stock exercise can contain several unrelated situations. Duplicate item codes may describe the same component. A repaired unit may be waiting to re-enter stock. A spare may be dedicated to a machine transferred to another site. Another may be an insurance spare deliberately held against a rare but disruptive failure.

AI can help assemble these relationships from the equipment register, purchasing records and maintenance history. The first useful output is a candidate classification with the supporting records and unresolved questions.

Automatic disposal is a different level of authority. Do not give it to a tool merely because it can rank a spreadsheet.

Start with the link between item and installed equipment. Check whether the machine is still operating and whether modifications changed the required component. A clean catalogue description can be misleading after a retrofit.

Ask what happens if the part is unavailable

For each candidate, examine the operational consequence of waiting. Can another machine take the load? Can the component be repaired locally? Is there an approved substitute? What is the current replenishment lead time, and has the supplier confirmed it?

The old purchase order is evidence of a previous transaction. It is a weak basis for assuming the same delivery promise today.

The inventory decision should combine this exposure with holding cost, condition and obsolescence. A component that cannot pass inspection is not useful protection, even if the system counts it as available. Conversely, an unused but serviceable spare may be fulfilling its intended role.

Keep safety-critical equipment and mandatory stock requirements within the site's engineering and compliance rules. A consumption model cannot waive those obligations.

Inspect the repair loop

A surprising amount can hide between “removed from machine” and “available for issue”. Consider a repairable drive. The replacement was fitted, the failed drive was shipped to a workshop, and the warehouse system still shows the old unit in a location nobody checks.

An assistant can reconcile the repair order, shipment, workshop response and expected return. It can flag an overdue answer or a serial-number mismatch. Someone still needs to confirm the repaired unit's acceptance and physical location.

Without that reconciliation, a purchasing recommendation may buy another unit while one sits at a supplier. Or it may refuse a purchase because an unusable unit appears in stock.

This is a useful place to start when historical consumption is too sparse for a meaningful demand forecast. Resolving an item’s actual status may be more valuable than predicting its next issue date.

Give the review a decision record

For each reviewed item, retain the equipment dependency, verified supply options, condition evidence and the person authorised to decide. Record the chosen action: retain, inspect, repair, transfer, consolidate or dispose through the approved process.

Include a reason and a review trigger. Retirement of the machine, an approved substitute or a changed supplier lead time may justify revisiting a decision that was sensible when made.

Avoid measuring the exercise only by inventory value removed. Track emergency purchases, downtime associated with unavailable materials and items found unusable when needed. Otherwise, a successful stock-reduction campaign can quietly transfer cost and risk into operations.

You do not need to classify the whole warehouse first. Choose a small group of expensive slow movers whose equipment links can be checked. Have maintenance, stores and purchasing examine the same records. Finance can then assess the economic trade-off with fewer hidden assumptions.

Bring that inventory sample to a Strategy Session. We can identify what can be reconciled automatically and what requires a physical or engineering check before a decision.