Lead Time Variability and Reorder Point Adjustments

Your Quoted Lead Time Is a Lie

Your supplier says 14 days. Your last three orders took 22, 19, and 28 days. So which number is in your reorder formula?

If you use 14, you'll stock out. If you use 28, you'll tie up cash. The answer is somewhere in between — and it depends on how variable your supplier actually is.

How to Measure Lead Time Variability

For every PO you've placed, log:

• The date you placed it
• The date you confirmed it with the supplier
• The date it shipped
• The date it arrived

The standard deviation of those arrival times is your lead time variability. The higher it is, the less you can trust your supplier's quote.

Adjusting Safety Stock for Variability

The practical way to size safety stock is in days of demand, not a statistics exam. Pick the number of days of buffer you want to hold before a SKU is "at risk," then let the system convert that into units using your actual forward demand — not a flat average.

The buffer should track your supplier's reliability:

• Reliable supplier (low lead-time variability): a tighter buffer — say 7 days — frees up cash.
• Unreliable supplier (high variability): widen the buffer — 14 days or more — and accept the carrying cost to avoid the stockout.

Because the conversion sums real day-by-day forecasted demand, a 10-day buffer on a SKU that sells heavier on weekends holds more units than one that sells flat — the buffer reflects how the product actually moves.

Dynamic Reorder Points

Static reorder points get stale fast. A dynamic reorder point recalculates every night based on:

• Latest sales velocity per SKU
• Latest lead-time average from your real PO history
• A forward-looking demand forecast, simulated day by day

This is what separates an inventory tool from a spreadsheet. Spreadsheets give you a number once a week. A real system gives you a fresh number every morning.

Try This Tomorrow

Pull the last 10 POs for your top 5 SKUs. Calculate the average and standard deviation of their lead times. If the standard deviation is more than 25% of the average, you have a supplier reliability problem worth solving.