Problem:

Data collected on the yearly registration for a six-sigma seminar at the Quality College
are shown in the following table: Show your computation and enjoy.
a. Develop a 3-year moving average to forecast registrations from year 4 to
year 12.
b. Estimate demand again for years 4 to 12 with a weighted moving average
in which registrations in the most recent years are weighted of 2 and
registrations in the other 2 years are each given a weight of 1.
c. Which of the two forecasting methods seems better?

year Registration Moving Average Weighted moving

average
1 4000
2 6000
3 4000
4 5000
5 10000
6 8000
7 7000
8 9000
9 12000
10 14000
11 15000
12

TYPE YOUR SOLUTION HERE USING EQUATION BUTTON

Solve for moving average:

Moving average=
∑ demand ∈ previous periods
n
4000+6000+ 4000
For year 4: =4666.6667
3
6000+4000+5000
For year 5: =5000
3
4000+5000+10000
For year 6: =6333.3333
3
5000+10000+8000
For year 7: =7666.6667
3
10000+ 8000+7000
For year 8: =8333.3333
3
8000+7000+9000
For year 9: =8000
3
 7000+9000+12000
For year 10: =9333.3333
3
9000+12000+14000
For year 11: =11666.6667
3
12000+ 14000+15000
For year 12: =13666.6667
3
16000

14000 yea
Registration Moving Average
r
12000 1 4000
2 6000
10000
3 4000
4 5000 4666.6667
8000
5 10000 5000
6000 6 8000 6333.3333
7 7000 7666.6667
4000 8 9000 8333.3333
9 12000 8000
2000 10 14000 9333.3333
11 15000 11666.6667
0 12 forecast 13666.6667
Year Year Year Year Year Year Year Year Year
4 5 6 7 8 9 10 11 12

Solve for weight moving average:

weight moving average=
∑ (weight for period n)(demand∈ period n)
∑ weights

weights applied
last year 2
two years ago 1
three years ago 1
sum of weights 4

(4000× 1)+(6000 ×1)+(4000 × 2)

For year 4: =4500
4
( 6000× 1 )+ ( 4000 ×1 ) +(5000× 2)
For year 5: =5000
4
 (4000× 1)+(5000× 1)+(10000× 2)
For year 6: =7250
4
(5000 ×1)+(10000 ×1)+(8000 × 2)
For year 7: =7750
4
(10000 ×1)+(8000 ×1)+(7000 × 2)
For year 8: =8000
4
(8000 ×1)+(7000 ×1)+(9000× 2)
For year 9: =8250
4
(7000 ×1)+(9000 × 1)+(12000× 2)
For year 10: =10000
4
(9000 ×1)+(12000 ×1)+( 14000× 2)
For year 11: =12250
4
(12000 ×1)+(14000 ×1)+(15000 ×2)
For year 12: =14000
4

16000

14000
yea Weighted
Registration
r
12000 moving average
1 4000
2
10000 6000
3 4000
8000
4 5000 4500
5
6000 10000 5000
6 8000 7250
7
4000 7000 7750
8 9000 8000
2000
9 12000 8250
100 14000 10000
11 Year 15000
Year Year Year Year 12250
Year Year Year Year
4 5 6 7 8 9 10 11 12
12 forecast 14000
Weighted moving average
 year Registration Moving Average Weighted moving
average
1 4000
2 6000
3 4000
4 5000 4666.6667 4500
5 10000 5000 5000
6 8000 6333.3333 7250
7 7000 7666.6667 7750
8 9000 8333.3333 8000
9 12000 8000 8250
10 14000 9333.3333 10000
11 15000 11666.6667 12250
12 13666.6667 14000

16000

14000

12000

10000
registration

8000

6000

4000

2000

0
Year 1 Year 2 Year 3 Year 4 Year 5 Year 6 Year 7 Year 8 Year 9 Year 10 Year 11 Year 12
period

Registration Moving Average Weighted Moving Average

Weighted moving average is better because it gives more weight to the recent
values than to the previous values. Therefore, it reacts more quickly to registration
changes. During the periods of increase (year 10 & year 11), it more closely tracks the
demand.