MGT 613 Assignment number 1 Spring 2023(Production & Operations Management)100% free download

 

 MGT 613 Assignment number 1 Spring 2023(Production & Operations Management)100% free download

Requirements:

1. Based on the given data, compute the exponential smoothing forecasts to develop a series of forecast for the period 2 - 12. Use value of smoothing constant a = 0.10

Based on the given data, let's compute the exponential smoothing forecasts using a smoothing constant (a) of 0.10 for periods 2-12.

Exponential Smoothing Forecast Calculation:

Calculations:

Don t Copy Paste

a

= 0.1

1 - a = 0.9

We'll use the following formula to calculate the exponential smoothing forecasts:

Forecast for period + = a x (Actual demand for period t) + (1 - a) × (Forecast

for period t-1)

Forecast for period 2 = is same as period 1 that is 42

Forecast for period 3 = (0.1 × 40) + (0.9 × 42) = 41,8

Forecast for period 4 = (0.1 × 43) + (0.9 × 41.8) = 41.92

Forecast for period 5 = (0.1 × 40) + (0,9 × 41.92)= 41.728

Forecast for period 6= (0.1 × 41) + (0.9 × 41.728)= 41,6552

Forecast for period 7= (0.1 × 39) + (0.9 × 41.6552)= 41.38968

Forecast for period 8= (0.1 × 46) + (0.9 × 41.38968)= 41850712

Forecast for period 9= (0.1 × 44) + (0.9 × 41.850712)= 42.0656408

Forecast for period 10= (0.1 × 45) + (0.9 × 42.0656408)= 42.3590767

Forecast for period 11= (0.1 × 39) + (0.9 × 42.3590767)= 41.923169

Forecast for period 12= <0.1 × 40) + (0.9 × 41.923169)= 41.7308521

 

2. What could be the possible range of values for smoothing constant (a) to be used in calculation forecast errors?

Range of Values for Smoothing Constant (a):

The range of values for the smoothing constant (a) typically falls between

O and 1. A smaller value of a (close to O) puts more weight on past observations, resulting in a smoother forecast that reacts slowly to changes in the data. On the other hand, a larger value of a (close to 1) puts more weight on recent observations, resulting in a forecast that reacts quickly to changes in the data

3. What should be the optimal value of smoothing constant in the prediction of forecast errors? Also, state when it is appropriate to use lower values of smoothing constant (a) and higher values of smoothing constant (a) Optimal Value of Smoothing Constant: The optimal value of the smoothing constant (a) depends on the characteristics of the data and the specific forecasting problem. Generally, it is determined through a process called "forecast evaluation" or "model selection," where different values of a are tested, and the forecast accuracy is assessed. The value of that yields the lowest forecast error (e.g., mean squared error) is considered the optimal value for a given dataset and forecasting problem.

 

In practice, it is appropriate to use lower values of a (e.g., closer to 0) when the data is stable and there is not much variability or rapid changes. Higher values of a (e.g. closer to 1) are suitable when the data is volatile and there are frequent fluctuations or sudden shifts,

Keep in mind that the optimal value of a may vary for different datasets, and it is recommended to perform thorough analysis and evaluation to determine the best value in each case.