Spare parts, and the storeroom that decides your repair times

The single biggest influence on how long a repair takes is whether the part is on the shelf. Not the skill of the technician, not the diagnostic tools. Whether the part is there. That is what a maintenance storeroom is for, and most of them are full of the wrong things.

The two failures, and they happen together

Almost every storeroom has both of these at once, which is why "spend less on stock" and "we never have the part" are usually said in the same meeting.

• •Stockouts on the parts that stop the building or the line
• •Shelves full of parts for equipment that was removed years ago

Start from the equipment, not from the shelf

The right question is never "what should we stock". It is "what does this critical machine need, and what happens if we do not have it". Work down from the criticality ranking and the answers fall out.

• •What fails on this machine, from your own history
• •What is the lead time on each of those parts, in reality rather than on the catalogue
• •What does an hour of this machine being down cost
• •Is there another machine that shares this part

Long lead time plus high consequence is the definition of a critical spare, whatever it costs. Short lead time plus low consequence is something you buy when you need it, however cheap it is.

Hold the part, not the assembly, unless time is the constraint

A seal kit is cheap and small. A complete pump is expensive and takes a shelf. Hold the assembly when the machine is critical enough that the repair has to be a swap and the diagnosis can happen afterwards. That is a deliberate and defensible choice, and it is a different decision from stocking spares in general.

Attach the parts list to the asset

The question during a breakdown is never "what is in stores". It is "what does this machine need and do we have it". If the parts list hangs off the asset record, that takes ten seconds. If it does not, somebody walks to the storeroom and looks at shelves.

Min and max, set from real usage

A minimum that triggers a reorder and a maximum that caps the holding. Set the minimum from usage during the lead time plus a margin for the consequence, and review both annually against what actually moved.

Count what moves, not everything

A full stock count once a year is a large effort producing a snapshot that is wrong within a week. Cycle counting, a few items every week with the fastest-moving and most critical counted most often, gives better accuracy for less work and finds errors while they are still small.

Deal with the dead stock, properly

Every storeroom holds parts for equipment that no longer exists. Nobody removes them because writing off stock is uncomfortable and keeping it is free until somebody looks at the space. Run the comparison once a year: any part with no movement in three years, cross-checked against the asset register, is a candidate to go.

Kitting is the highest-return habit

Picking the parts for scheduled jobs the day before, into a box with the job number on it, removes the single biggest interruption in a technician’s day. It costs a storeperson an hour in the morning and returns several hours of skilled time.

The measures worth having

• •Stockout rate on critical parts, which is the only service measure that matters
• •Jobs delayed waiting for parts, counted and trended
• •Stock value, split into critical, consumable and dead
• •Count accuracy from cycle counting