Step 1: Understand the concept of forecast error in different forecasting methods.
Forecast error is the difference between the actual demand and the forecasted demand for a given period. Some forecasting methods adjust the next forecast based on this error to make them more responsive to changes.
Step 2: Evaluate each forecasting method.
(1) Simple average method: This method calculates the average of all past data to forecast the next period. It does not explicitly incorporate forecast error in an iterative manner for adjustment.
(2) Weighted moving average method: This method assigns different weights to historical data, with more recent data typically receiving higher weights. While it gives more importance to recent observations, it doesn't directly use a fraction of the forecast error from the previous period to adjust the next forecast.
(3) Moving average method: This method calculates the average of demand over a specific number of recent periods. It smooths out fluctuations but does not incorporate a fraction of the forecast error to adjust the subsequent forecast.
(4) Exponential smoothing method: Exponential smoothing is a forecasting method that computes the next forecast by taking the previous period's forecast and adjusting it by a fraction of the previous period's forecast error. The formula for simple exponential smoothing is:
\[
F_{t+1} = F_t + \alpha (A_t - F_t)
\]
Where:
\( F_{t+1} \) = Forecast for the next period
\( F_t \) = Forecast for the current period
\( A_t \) = Actual demand for the current period
\( (A_t - F_t) \) = Forecast error for the current period
\( \alpha \) = Smoothing constant (a fraction between 0 and 1)
Here, \( \alpha \) is the "fraction of forecast error" that is taken into account for the next period's forecast.
Step 3: Conclude the method that uses a fraction of forecast error.
Exponential smoothing explicitly uses a fraction of the forecast error to adjust the forecast for the next period.
The final answer is $\boxed{\text{4}}$.