Every maintenance storeroom has two problems at once. There's too much money tied up in parts nobody's touched in years, and there's a stockout on the one part that would have saved a 14-hour outage.
Both come from the same place. Nobody decided, part by part, why it's on the shelf. Parts got stocked because a vendor recommended a spares kit, or because someone got burned once, or because the machine came with a box of them.
The decision is two costs
For any spare, you're comparing the cost of not having it when you need it with the cost of keeping it until you do.
The first is roughly downtime cost per hour times the hours you'd wait, and the wait is mostly lead time. So each part needs a downtime cost, a lead time and a carrying cost.
Downtime cost is what it costs per hour when the equipment that uses the part is down. A packaging line that feeds every shipment might cost $8,000 an hour in lost output and catch-up overtime. A backup compressor might cost nothing while the primary runs. Ask production and finance together. You'll get two different numbers, so use the one you can both defend.
Lead time means the real one, not the catalog's. Look at your last few purchase orders, or call the supplier and ask what an emergency order takes today. For OEM parts on older or imported equipment, I often see quotes of 6 to 16 weeks. Check whether a local distributor stocks it, too. A part you can get across town in two hours is a different decision from one coming from overseas.
Carrying cost is usually figured at 20 to 30 percent of part value per year. That covers capital, space and the risk the part goes obsolete or degrades. Seals and belts do age on the shelf, and some electronics have a shelf life.
Rank equipment before parts
You don't need to run numbers on all 8,000 items. Rank the equipment first, then work down. I use three levels:
- A: failure stops production or creates a safety or environmental risk, and there's no backup.
- B: failure slows production, or there's a workaround that costs real money.
- C: failure is an inconvenience, or there's a redundant unit.
For the parts on A equipment, ask whether that part failing takes the machine down. A bearing on the main drive does. A cracked guard cover probably doesn't, even on the most important machine in the plant.
The parts that live on A equipment and stop it when they fail are your critical spares. In most plants I visit, that's a few hundred items, not thousands.
Run the comparison
For each critical spare:
- Cost of a stockout = downtime cost per hour × hours to get and install the part
- Cost of stocking = part cost × carrying rate × years you expect to hold it
A $3,000 gearbox on a line that costs $5,000 an hour, with a four-week lead time, isn't a hard call. One stockout costs more than carrying it for a decade.
A $40,000 spindle with a two-day turnaround from a local rebuilder is a different story. A standing agreement with the rebuilder might beat a spindle on the shelf.
If your maintenance records show a part fails every 18 months, use that. If you have no history, use the manufacturer's guidance or what plants with the same equipment see, and revisit after a year of tracking your own.
Min/max for parts that rarely move
Standard reorder math assumes steady demand, and critical spares don't have it. A part might sit for three years and then get used twice in a month, so I set these by judgment:
- For most critical spares, min 1 and max 1. Use one, reorder one.
- For parts installed in several places, like the same motor on six conveyors, min might be 1 or 2 depending on lead time and how many could fail together.
- For parts that fail in pairs or sets, stock the set.
The reorder has to fire on the first use. If your system flags anything with no movement in 12 months as a candidate for zero stock, exclude critical spares from that rule. I've seen automated slow-mover cleanups remove exactly the parts that mattered most.
Count critical spares on a fixed schedule, at least quarterly, and look at their condition while you're there. A spare the system says is on the shelf but isn't is worse than having none, because everyone relaxes. Look for rust and torn packaging, and for a new part that got quietly swapped for a used one.
Parts for machines you don't own anymore
This is the easy money. When a machine is retired, its spares usually stay in the storeroom forever, because nobody connects the decommission to the parts list. I've seen storerooms where 15 to 25 percent of the value was parts for equipment no longer in the building.
To find them:
- List every machine retired, sold or scrapped in the last ten years.
- Pull each one's parts list (bill of materials or where-used).
- Check whether any of those parts fit equipment still in service.
- Anything that doesn't is a candidate to return, sell or scrap.
Call the original vendor first, since some take back unused parts for credit. Surplus dealers will buy others. Even scrapping frees shelf space. Then add a step to your decommissioning checklist so the pile doesn't rebuild.
Where to start
Pick your five most important A machines. Build their critical spares lists, check the counts and set min/max. That covers most of your real outage risk, and you can work through the rest of the plant a line at a time.