Service operation management - Assignment Example

r Distribution Centre 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Darwin 18 15 15 13 12 12 14 13 13 13 12 10 11 11.8 11.56 11.27 Brisbane 22 25 20 22 21 20 23 20 20 21 23 20 21 21 21.2 21.24 Sydney 48 47 49 52 53 53 55 56 58 61 63 66 68 63.2 64.24 64.89 Adelaide 36 38 39 42 38 41 42 44 45 47 47 51 52 48.4 49.08 49.5 Hobart 18 20 22 18 22 19 18 20 21 19 21 20 18 19.8 19.56 19.67 b. Demand Forecasting using Exponential Smoothing Model is shown below. Table1.3 shows the forecasting model for alpha as .

2 while Table 1.4 shows the forecasting model for alpha as .4. .   Week Number Distribution Centre   1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Darwin Actual 18 15 15 13 12 12 14 13 13 13 12 10 11       Forecast 16 16.8 16.08 15.65 14.59 13.55 12.93 13.36 13.22 13.13 13.08 12.65 11.59 11.35 11.21 11.13 Brisbane Actual 22 25 20 22 21 20 23 20 20 21 23 20 21       Forecast 22.33 22.2 23.32 21.99 22 21.6 20.96 21.77 21.06 20.64 20.78 21.67 21 21 21 21 Sydney Actual 48 47 49 52 53 53 55 56 58 61 63 66 68       Forecast 48 48 47.8 48.04 48.83 49.67 50.33 51.27 52.21 53.37 54.9 56.52 58.41 62.25 64.55 65.93 Adelaide Actual 36 38 39 42 38 41 42 44 45 47 47 51 52       Forecast 37.67 37.33 37.47 37.77 38.62 38.49 39 39.6 40.48 41.38 42.51 43.4 44.92 47.75 49.45 50.

47 Hobart Actual 18 20 22 18 22 19 18 20 21 19 21 20 18       Forecast 20 19.6 19.68 20.14 19.72 20.17 19.94 19.55 19.64 19.91 19.73 19.98 19.99 19.19 18.72 18.43 c. Exponential smoothing is generally preferred over moving average forecasting method because it takes into account both the actual values and the previous forecasts. Also the mean square error received in the case of exponential smoothing is generally less as compared to moving average method. Between the two moving average methods, the 5 week moving average forecasting generally proves to be more accurate than the 3 period moving average forecasting because it takes into account large historical data.

Among the exponential smoothing methods, the optimal value of alpha is generally found out by minimizing the mean square error through excel solver. d. Besides past demand several other factors might come into play in forecasting future demand. These factors could range from local to global. Local factors may include current market share of SB Coffee in each distribution centre, population of each geography,