OptimalSPARES™ Wiki / Criticality
Method
Spares criticality
The decision to stock a part, and to what service level, is driven by the consequence of not having it, not by what the part costs. A cheap gasket on a critical pump can matter more than an expensive spare on a standby unit. Criticality is how that is made explicit.
What makes a spare critical
| Consequence of stockout | The downtime, safety or environmental cost if the asset waits for the part. The dominant factor, inherited from the parent asset criticality. |
|---|---|
| Lead time | How long to get another one. A long or unreliable lead time raises the case to hold, because you cannot buy your way out quickly. |
| Demand rate and predictability | How often it is used and how steady that is. Rare and erratic demand is the hard case, and the reason slow-mover models exist. |
| Number of installed units | How many are in service across the site, which drives the aggregate failure rate and the pooling opportunity. |
| Alternatives | Whether a substitute, a repair or a local supplier exists. A part with no alternative is more critical than its price suggests. |
| Cost and shelf life | Purchase and holding cost, and whether it degrades or becomes obsolete on the shelf. |
From criticality to service level
Criticality is banded, and each band is given a target service level, which is the probability of not stocking out in a cycle. The stocking maths on the next page then sizes the safety stock to hit it:
| Critical | Stockout stops a critical asset or carries a safety or environmental consequence. Target a high service level, around 98 to 99% or more. |
|---|---|
| Essential | Stockout hurts but can be worked around for a time. A middle target, around 95%. |
| Routine | Low consequence, easily obtained. A lower target, around 90%, or stock only to a min. |
Insurance spares are different
An insurance spare, a large transformer or a bespoke rotor, may have essentially no demand history and yet be held, because a failure would be catastrophic and the lead time is measured in months or years. These are stocked on a risk case, comparing the holding cost against the expected downtime cost, not on an economic order quantity. OptimalSPARES™ flags and reasons about these separately, so they are neither ignored nor stocked by a formula that does not fit.
Where OptimalSPARES™ fits
OptimalSPARES™ takes the asset criticality from the reliability model in OptimalAvailability Studio™, combines it with lead time, demand and alternatives into a spare criticality, and turns that into the service-level target that drives the stocking policy. The reasoning is recorded, so a stocking decision can be explained and defended rather than being a number someone once typed. The scoring is AI-assisted and applied consistently across the whole estate, so criticality is comparable site to site rather than depending on who did each one.