Balance orders, inventory, and capacity across changing periods. See smoothed demand, safety stock, and trends. Make calmer planning choices with transparent metrics and charts.
Use this sample to test the calculator and compare how smoothing stabilizes volatile demand.
| Period | Actual Demand | Forecast Used | Updated Smoothed Value |
|---|---|---|---|
| 1 | 120 | 120 | 120.00 |
| 2 | 135 | 120 | 125.25 |
| 3 | 128 | 125.25 | 126.21 |
| 4 | 140 | 126.21 | 131.04 |
| 5 | 150 | 131.04 | 137.68 |
| 6 | 145 | 137.68 | 140.24 |
Updated smoothed value: St = α × At + (1 − α) × Ft
Forecast used for period t: Ft = St−1
Next period forecast: Ft+1 = St
Error: Error = Actual − Forecast
MAE: Mean of absolute errors
RMSE: Square root of mean squared errors
MAPE: Mean of absolute percentage errors × 100
Coefficient of variation: CV = Standard Deviation ÷ Mean × 100
Reduction: ((CV actual − CV smoothed) ÷ CV actual) × 100
Safety stock: Z × σ × √Lead Time
Lead time demand: Next Forecast × Lead Time
Reorder point: Lead Time Demand + Safety Stock
Inventory position: Current Inventory + On Order − Backorders
Raw order gap: max(0, Reorder Point − Inventory Position)
Smoothed order: Dampening × Raw Gap + (1 − Dampening) × Next Forecast
It measures how a smoother forecast reduces demand noise before planning replenishment. The calculator compares raw variability and smoothed variability to show whether ordering signals become calmer and easier to manage.
Use a higher alpha when demand shifts quickly and older observations lose value fast. It makes the forecast more responsive, but it also allows more volatility into the planning signal.
The service factor controls how much protection you want against uncertainty during lead time. Higher values increase safety stock and reduce shortage risk, but they also raise inventory exposure.
It shows how much the coefficient of variation falls after smoothing. A higher percentage means the forecast is less erratic than the original demand history, which can lower overreaction in replenishment planning.
No. It is a strong planning aid, not a complete policy engine. Seasonality, supplier disruptions, promotions, order batching, and multi-echelon constraints may still require separate business rules.
At least three periods are required, but more history usually improves stability. Use consistent time buckets such as weeks or months so the smoothing logic compares like with like.
This file uses basic exponential smoothing, which is best for level demand with noise. For strong seasonality or trend, consider Holt or Holt-Winters methods and compare their error metrics separately.
The dampening factor intentionally softens large swings by blending the raw gap with the smoothed forecast. This supports steadier replenishment and helps reduce the bullwhip effect across periods.
Important Note: All the Calculators listed in this site are for educational purpose only and we do not guarentee the accuracy of results. Please do consult with other sources as well.