Most sites do not have a critical-spares philosophy. They have an inventory that accreted: spares bought with projects and never rationalized, spares added after each painful failure, and spares missing precisely where consequence and lead time are worst. The result is simultaneously too much capital on shelves and too much exposure on the assets that matter.
Consequence and lead time, not habit, define criticality
A spare earns the word critical when two things are simultaneously true: the failure it protects against stops or materially constrains the operation, and the replacement cannot be obtained within a tolerable outage window. Everything else is a stocking decision, not a criticality decision. This framing immediately reorders most inventories — commodity items held in bulk lose their claim, while long-lead engineered components with no interchangeable substitute rise to the top.
Credibility of failure keeps the list honest
Consequence alone over-buys. The third input is whether the failure mode is credible on this asset in this service: condition history, known degradation mechanisms, fleet experience, and the quality of monitoring that would give warning. A component with severe consequence, a three-week lead time, and strong condition monitoring may justify a supplier agreement instead of a shelf. A modest component with an eighteen-month lead time and a silent failure mode may justify two on the shelf.
Mitigation is broader than buying
The cost of the philosophy stays controlled because stocking is only one of its tools. Reservation agreements with suppliers, shared fleet or sister-site pools, refurbishment paths for repairable components, qualified alternates documented before the emergency, and engineering assessments that extend tolerable outage windows all reduce exposure without adding inventory. The philosophy decides which tool each risk gets.
- Rank by consequence and lead time; filter by failure credibility
- Document the tolerable outage window each critical spare protects
- Use supplier agreements, pooling, and qualified alternates before shelf stock where they genuinely cover the risk
- Review the list on operating-profile changes, not just on a calendar
The goal is not a bigger warehouse or a smaller one. It is knowing, for every consequential failure, exactly how the organization intends to survive it.