Demand forecasting
Croston's Method
By the Supply Chain Manager team · Updated October 16, 2026
Croston's method forecasts intermittent demand by smoothing two things separately: the size of non-zero orders and the interval between them.
The forecast demand rate is the smoothed order size divided by the smoothed interval. Estimates only update in periods with demand, so a run of zero weeks does not drag the size estimate down.
The Syntetos-Boylan Approximation (SBA) is a common variant that corrects a known upward bias in the original method by multiplying the result by (1 − α/2).
How to calculate Croston's method
Demand Rate = Smoothed Order Size ÷ Smoothed Interval Between Orders SBA variant: Demand Rate × (1 − α ÷ 2)
Croston's method example
Illustrative numbers: orders average 12 units and arrive about every 4 weeks. Croston's forecast is 12 ÷ 4 = 3 units a week. With α = 0.1, the SBA version is 3 × 0.95 = 2.85 units a week.
Why Croston's method matters for distributors
For spare parts and specialist lines it gives a more realistic demand rate than an average, which feeds better safety stock and reorder decisions.
Common Croston's method mistakes
- Using Croston's method on fast, steady SKUs, where simpler methods do better.
- Reading the per-period rate as the quantity you will sell each period.
Croston's method in Supply Chain Manager
Supply Chain Manager includes classical Croston and Croston SBA in its per-SKU model backtests.
Open the live demoExamples on this page use made-up round numbers to show the calculation. They are not benchmarks or results from real businesses.