Safety stock calculator
The cushion that covers bad weeks — sized from how your usage and lead times actually vary, not from a guess.
Covers the worst observed cycle (120 units) minus a normal one (56). Your reorder point becomes 56 + 64 = 120.
The formula
Safety stock = (max daily usage × max lead time) − (avg daily usage × avg lead time)
A normal replenishment cycle consumes average usage × average lead time. A bad cycle — busy week, slow vendor, or both at once — consumes up tomax × max. Safety stock is the difference: the units that keep a bad cycle from becoming a stockout.
Where the inputs come from
Pull them from history, not memory: your busiest week's daily usage over the last couple of quarters, and the slowest the vendor has actually delivered. If consumption isn't logged anywhere, that's the real gap — a movement ledger makes these numbers a query instead of an argument. (Until then, the inventory template starts the habit.)
The statistical version
With clean data, safety stock can be sized to a service level:Z × σ(demand over lead time), where Z is 1.65 for ~95% service. It's more precise and more brittle — garbage variance in, confident garbage out. The max/average method above is the honest choice for most shops' data quality, and it fails conservative.
Use it with the reorder point
Safety stock is the cushion; the reorder point is the tripwire that protects it. Compute safety stock here, add average-cycle demand, and that's the level where the PO should fire — automatically, if something is watching.