Answer Demand Forecasting & ABC Analysis (Exercise 2)
Answer Demand Forecasting & ABC Analysis (Exercise 2)
ABC Analysis
The Answer of Exercise II
Single Moving Average (3 & 5 Months)
3 Months Single Moving Average 5 Months Single Moving Average
Demand (1.000
Month
Kg)
Forecast Error MAD MSE Forecast Error MAD MSE
Jan 20
Feb 21
Mar 19
Apr 17 20 -3 3 9
Mei 22 19 3 3 9
From the data of MAD and MSE above the 5 month single moving averages
give better forecast than 3 month single moving averages.
Double Moving Average
Period Demand 4 Years Single Moving Average 4 Years Double Moving Average a Value b Value Forecast
1 120
2 125
3 129
4 124 124.50
5 130 127.00
6 140 130.75
Feb 21 20 1 1 20 1 1 20 1 1
Mar 19 20.1 1.1 1.21 20.5 1.5 2.25 20.9 1.9 3.61
Apr 17 19.99 2.99 8.94 19.75 2.75 7.56 19.19 2.19 4.80
Mei 22 19.69 2.31 5.33 18.38 3.63 13.14 17.22 4.78 22.86
Juni 24 19.92 4.08 16.63 20.19 3.81 14.54 21.52 2.48 6.14
Juli 18 20.33 2.33 5.43 22.09 4.09 16.76 23.75 5.75 33.09
Agst 21 20.10 0.90 0.82 20.05 0.95 0.91 18.58 2.42 5.88
Sept 20 20.19 0.19 0.03 20.52 0.52 0.27 20.76 0.76 0.57
Oct 23 20.17 2.83 8.02 20.26 2.74 7.50 20.08 2.92 8.55
Nov 22 20.45 1.55 2.40 21.63 0.37 0.14 22.71 0.71 0.50
Des 19 20.61 1.61 2.58 21.82 2.82 7.93 22.07 3.07 9.43
Each α give different MAD and MSE as shown in previous table. When detailing α value and real demand in each
month show that α=0.90 give nearest forecast to lower and highest real demand value 17 and 24, α=0.50 give
nearest forecast to 22 and 23, and others give nearest forecast to α=0.10. In conclusion, the lowest α value give
the higher weight or lowest error than others.
Single Exponential Smoothing
Double Exponential Smoothing
Period Demand Single Exponensial Smoothing Double Exponensial Smoothing a Value b Value Forecast
90.0%
80.0%
70.0%
60.0%
50.0% Series1
40.0%
30.0%
20.0%
10.0%
0.0%
1 2 3 4 5 6 7 8 9 10
Data on sales and advertising dollars for the past six months are shown below.