Test: Forecasting Level - 1 - Mechanical Engineering MCQ

Choosing the 'n' Value in a Simple Moving Average Method for Stable Demand

- When utilizing a simple moving average method to forecast demand or trends, especially in scenarios where the demand is stable, the selection of the 'n' value, which represents the number of periods to be averaged, is crucial. The correct choice of 'n' influences the smoothness of the resulting moving average and how well it can indicate underlying trends without being overly sensitive to short-term fluctuations.

- Why a Higher 'n' Value is Recommended for Stable Demand:

- Smoothens Random Fluctuations: A higher 'n' value averages out data over a longer period, which effectively smoothens out random fluctuations and noise in the data. This is particularly beneficial in stable demand scenarios where you're interested in identifying the long-term trend rather than reacting to short-term variations.

- Provides a More Stable Trend Line: In environments where demand is generally stable, a higher 'n' value yields a moving average that is less reactive to minor changes. This results in a trend line that more accurately reflects the stable nature of the demand, making it easier for businesses to plan and make decisions.

- Reduces the Impact of Anomalies: Using a higher 'n' value dilutes the impact of any anomalous or outlier data points, which might otherwise skew the analysis and lead to incorrect interpretations or predictions.

- Better for Long-term Planning: For businesses or analyses focused on long-term planning and forecasting, a higher 'n' value in a simple moving average method provides a forecast that aligns better with these goals by highlighting more enduring trends rather than ephemeral changes.

- Considerations:

- While a higher 'n' value is generally beneficial in stable demand scenarios, it's important to note that too high a value can make the moving average too insensitive to genuine changes in the trend. Therefore, the specific choice of 'n' should be based on the industry, the typical cycle lengths observed in the demand, and the strategic focus of the forecasting effort (short-term vs. long-term).

- Additionally, the choice of 'n' might need to be adjusted as more data becomes available or if the nature of the demand changes (e.g., if the market becomes more volatile).

Regression analysis is a powerful method for forecasting when a demand pattern is consistently increasing or decreasing. By analyzing historical data and identifying the underlying trend, regression analysis can provide more accurate and robust forecasts compared to simpler methods. It is important to properly implement and interpret regression analysis to ensure reliable forecasts for future demand levels.

= 94.5

F₄ = αd₃ + (1 − α)F₃

F₃ = (0.2)(480) + (0.8)500

= 96 + 400 = 496

α = 0.2, D_March = 20 units D_April = 25 D_May = 26

F_Mar = 20 units F_April = 20 F_May = 21 F_Jun =?

F_April = α × D_Mar + (1 − α) F_Mar = 0.2 × 20 + 0.8 × 20 = 20

F_May = α × D_April + (1 − α) × F_April = 0.2 × 25 + 0.8 × 20 = 21

F_June = α × D_May + (1 − α) × F_May

F_May = 0.2 × 26 + 0.8 × 21 = 22 units

For stable demand ‘n’ value should be more

⇒ α will be less.

F_t+1 = αD_t + α(1 − α)D_t−1 + α(1 − α)²D_t−2+ α(1 − α)³D_t−3

= 0.2(115) + 0.2(1 − 0.2) × 108

+ 0.2(1 − 0.2)2 × 98

+ 0.2(1 − 0.2)3 × 102

= 63.26

F_t+1 = F_t + α(D_t − F_t )

F_July = F_June + α(D_June − F_June )

F_July = 80 + 0.28(90 − 80)

= 82.8

The demand is having an increasing pattern. So use regression analysis