First-Order Exponential Smoothing When you report to your...

First-Order Exponential Smoothing When you report to your supervisor that the moving average forecasting technique is not suitable for the product, he remembers that your colleague in charge of forecasting had been working on introducing the first-order exponential smoothing technique for this product. Therefore, your supervisor gives you the information in Figure 9.8.3.1, showing the demand for the product (January to October) and the forecast using the first-order exponential smoothing technique with $\alpha=0.3$ of the product (January to July). $$ \begin{array}{|l|c|c|c|c|c|c|c|c|c|c|} \hline & \text { Jan. } & \text { Feb. } & \text { Mar. } & \text { Apr. } & \text { May. } & \text { Jun. } & \text { July. } & \text { Aug. } & \text { Sept. } & \text { Oct. } \\ \hline \text { Demand } & 151 & 135 & 143 & 207 & 199 & 175 & 111 & 95 & 119 & 191 \\ \hline \text { Forecast } & 187 & 176 & 164 & 158 & 172 & 180 & 179 & & & \\ \hline \end{array} $$ To evaluate your supervisor's suggestion, you execute the following steps: a. Compute the forecast for August, September, and October and for the following month, November. b. Calculate the mean absolute deviation (MAD) for November assuming $\operatorname{MAD}(\operatorname{Jan})=18$ and the smoothing parameter $\alpha$. c. In the preceding exercise, could you have obtained a result comparable to the one for the parameter $\alpha$ calculated above by changing $\mathrm{n}$, that is, the number of observed values? d. Decide whether the chosen first-order exponential smoothing technique with parameter $\alpha$ calculated above is appropriate for this product. e. What can you say in general about the choice of $\alpha$ depending on the product life cycle?

First-Order Exponential Smoothing

When you report to your supervisor that the moving average forecasting technique is not suitable for the product, he remembers that your colleague in charge of forecasting had been working on introducing the first-order exponential smoothing technique for this product. Therefore, your supervisor gives you the information in Figure 9.8.3.1, showing the demand for the product (January to October) and the forecast using the first-order exponential smoothing technique with $\alpha=0.3$ of the product (January to July).
$$
\begin{array}{|l|c|c|c|c|c|c|c|c|c|c|}
\hline & \text { Jan. } & \text { Feb. } & \text { Mar. } & \text { Apr. } & \text { May. } & \text { Jun. } & \text { July. } & \text { Aug. } & \text { Sept. } & \text { Oct. } \\
\hline \text { Demand } & 151 & 135 & 143 & 207 & 199 & 175 & 111 & 95 & 119 & 191 \\
\hline \text { Forecast } & 187 & 176 & 164 & 158 & 172 & 180 & 179 & & & \\
\hline
\end{array}
$$
To evaluate your supervisor's suggestion, you execute the following steps:
a. Compute the forecast for August, September, and October and for the following month, November.
b. Calculate the mean absolute deviation (MAD) for November assuming $\operatorname{MAD}(\operatorname{Jan})=18$ and the smoothing parameter $\alpha$.
c. In the preceding exercise, could you have obtained a result comparable to the one for the parameter $\alpha$ calculated above by changing $\mathrm{n}$, that is, the number of observed values?
d. Decide whether the chosen first-order exponential smoothing technique with parameter $\alpha$ calculated above is appropriate for this product.
e. What can you say in general about the choice of $\alpha$ depending on the product life cycle?

Key Concepts

First-Order Exponential Smoothing

This concept involves a forecasting method that weights current and past observations using a smoothing constant, ?, to assign more significance to recent data. It is an effective technique for time series with no strong trend or seasonal pattern, and the basic idea is to update forecasts as new demands are observed, thus providing a balance between responsiveness and stability in the forecasts.

Mean Absolute Deviation (MAD)

MAD is a measure of forecast accuracy that calculates the average absolute error between the forecasted and actual values. It provides a straightforward metric to assess how well the forecasting model is performing, with lower MAD values indicating a closer match between predicted and actual demand. This metric is crucial for comparing different forecasting approaches or parameter settings.

Parameter (Alpha) Selection

The smoothing parameter, ?, in exponential smoothing dictates the weight given to the most recent observation relative to previous forecasts. A larger ? places more emphasis on recent data, making the forecast more responsive to changes, while a smaller ? results in a smoother forecast that is less sensitive to short-term fluctuations. Choosing the right ? is critical for achieving a balance between overreacting to random noise and underreacting to true shifts in the data.

Forecast Evaluation

Forecast evaluation involves the analysis of forecast errors using statistical measures such as MAD to determine the reliability and accuracy of a forecasting method. This process is essential to validate whether the chosen forecasting method and its parameters, like ? in exponential smoothing, are appropriate for the specific context or product. It allows forecasters to compare different models and refine methods based on historical performance.

Product Life Cycle Considerations

The choice of forecasting method and the tuning of parameters such as ? are often influenced by the stage of a product’s life cycle. For products in an early stage or a rapidly changing market, a higher ? may be preferred to quickly adapt to demand changes. Conversely, for mature or stable products, a lower ? might be appropriate to smooth out short-term volatility. Understanding the product life cycle is therefore essential in selecting and adjusting forecasting models.

Please give Ace some feedback