Calculate demand forecasts using a weighted moving average. Assign custom weights to recent periods for more responsive demand projections.
The Weighted Moving Average (WMA) forecast assigns different weights to each of the last N demand periods, allowing you to emphasize more recent data. Unlike a simple moving average that weights all periods equally, WMA lets you control how much influence each period has on the forecast.
Typically, the most recent period receives the highest weight, the next most recent gets a lower weight, and so on. The forecast equals the sum of each period's demand multiplied by its weight, divided by the sum of all weights.
This calculator lets you enter demand values and corresponding weights for up to six periods. It normalizes the weights and computes the weighted average forecast for the next period.
Supply-chain managers, warehouse operators, and shipping coordinators rely on precise weighted moving average data to maintain efficiency and control costs across complex distribution networks. Revisit this calculator whenever conditions change to keep your logistics plans aligned with real-world performance.
When demand is changing, giving more weight to recent periods produces a more responsive forecast than a simple moving average. This calculator makes it easy to experiment with different weight schemes to find the combination that best tracks your demand pattern. Real-time recalculation lets you model different scenarios quickly, ensuring your logistics decisions are backed by accurate, up-to-date numbers.
Forecast = Σ(w_i × D_i) / Σ(w_i) Where: w_i = Weight assigned to period i D_i = Demand in period i Σ(w_i) = Sum of all weights (used for normalization)
Result: Forecast = 113.3
Weighted sum = (100×1) + (110×2) + (120×3) = 100 + 220 + 360 = 680. Sum of weights = 1 + 2 + 3 = 6. Forecast = 680 / 6 = 113.3. The most recent period (120) has the most influence.
Common weight schemes include linear (1, 2, 3, ..., N), triangular, and custom patterns. Linear weights are the most popular because they are intuitive and provide a good balance between responsiveness and stability. More aggressive schemes concentrate weight heavily on the last 1–2 periods.
Both methods emphasize recent data, but they differ in implementation. WMA uses explicit weights for a fixed window. Exponential smoothing implicitly weights all historical data with exponentially decaying influence controlled by a single parameter. For most applications, exponential smoothing is preferred for its simplicity.
WMA is most useful when you have strong domain knowledge about how much each past period should influence the forecast. For example, if the last two months are highly relevant but earlier months are unreliable due to a known disruption, you can zero out their weights. This flexibility is WMA's key advantage.
SMA gives equal weight to all periods in the average. WMA assigns custom weights, typically giving more importance to recent periods. This makes WMA more responsive to demand changes than SMA.
Start with linearly increasing weights (1, 2, 3, ...) and test against historical data. Adjust to minimize forecast error. Some planners use exponentially increasing weights for faster responsiveness.
Typically 3 to 6 periods. Including too many periods dilutes the benefit of differential weighting. The optimal number depends on your demand pattern stability and forecasting horizon.
Exponential smoothing is generally preferred because it automatically applies exponentially decreasing weights with just one parameter (α). WMA requires choosing individual weights for each period, which is more flexible but harder to optimize.
WMA does not inherently handle seasonality. For seasonal demand, consider deseasonalizing the data before applying WMA, or use a seasonal forecasting method like Holt-Winters.
No. The calculator divides the weighted sum by the total weight, so any weight values work. Using integers like 1, 2, 3 is simpler than fractions like 0.167, 0.333, 0.500.