The following gives the number of pints of type B blood used at Woodlawn Hospital in the past 6 weeks:

WEEK OF PINTS
31-Aug 360
7-Sep 389
14-Sep 410
21-Sep 381
28-Sep 368
5-Oct 374

a) Forecast the demand for the week of October 12 using a 3-week moving average.
OCTOBER 12 FORECAST: 3-
WEEK MOVING AVERAGE
3-WEEK
PINTS MOVING
PERIOD AVERAGE
USED

31-Aug 360
7-Sep 389
14-Sep 410
21-Sep 381 386 (360 + 389 + 410)/3
28-Sep 368 393 (389 + 410 + 381)/3
5-Oct 374 386 (410 + 381 + 368)/3
12-Oct NEED 374 (381 + 368 + 374)/3

b) Use a 3-week weighted moving average, with weights of .1, .3, and .6, using .6 for the most recent week.
Forecast demand for the week of October 12.
OCTOBER 12 FORECAST: 3-
WEEK WEIGHTED MOVING AVERAGE Weights
0.1 3 wks. ago
PINTS 3-WEEK
PERIOD WEIGHTS 0.3 2 wks. ago
USED WMA
0.6 Last wk.
31-Aug 360 1 Sum of Weights
7-Sep 389
14-Sep 410
21-Sep 381 0.1
28-Sep 368 0.3
5-Oct 374 0.6
12-Oct DEMAND 1 373 [(381*.1)+(368*.3)+(374*.6)]/1

c) Compute the forecast for the week of October 12 using exponential smoothing with a forecast for August 31
of 360 and a = .2.
F1 = Ft-1 + a(At-1 - Ft-1)
Ft = new forecast PINTS SMOOTHING
PERIOD
USED FORECAST
Ft-1 = previous period forecast
a = smoothing (or weighting) constant (0 ≤ a ≤ 1) 31-Aug 360
At-1 = previous period actual demand 7-Sep 389 360
14-Sep 410 366 360 + .2 (389- 360)
21-Sep 381 375 366 + .2 (410 - 366)
28-Sep 368 376 375 + .2 (381 - 375)
5-Oct 374 374 376 + .2 (368 - 376)
12-Oct 374 374 + .2 (374 - 374)
360 + .2 (389- 360)
366 + .2 (410 - 366)
375 + .2 (381 - 375)
376 + .2 (368 - 376)
374 + .2 (374 - 374)
Refer to Problem 4.2. Develop a forecast for years 2 through 12 using exponential smoothing with a = .4 and a forecast for yea
your new forecast on a graph with the actual data and the naive forecast. Based on a visual inspection, which forecast is bette
NAÏVE EXPONENTIAL
YEAR DEMAND SMOOTHING
Prob. 4.2 FORECAST FORECAST
1 7 7 14
2 9 7 6.4
12
3 5 9 7.4
4 9 5 6.5 10
5 13 9 7.5
8
6 8 13 9.7
7 12 8 9.0 6
8 13 12 10.2 4
9 9 13 11.3
2
10 11 9 10.4
11 7 11 10.6 0
1 2 3 4 5 6 7 8 9 10 11 12 13
12 7 9.2

The Naive Approach is a good entry into quantitative forecasting, tracking all the ups and downs. However, based u
Trendline, which represents the yearly Demand volume, the Exponential Smoothing Approach flows closer to the lin
Approach uses a low valued smoothing constant (forecast error) of .4 to balance the random variations, this repres
stable, and provides a better forecasting result. Hence, the weighted forecast gives a better depiction of a good acc
 a = .4 and a forecast for year 1 of 6. Plot
tion, which forecast is better?

NAÏVE FORECAST
Linear (NAÏVE FORECAST)
EXPONENTIAL SMOOTHING
FORECAST

9 10 11 12 13

d downs. However, based upon the Naïve Forecast Linear

proach flows closer to the line. The Exponential Smoothing
ndom variations, this represents the average as being more
etter depiction of a good accurate forecast.
The Carbondale Hospital is considering the purchase of a new ambulance. The decision will rest partly on the
anticipated mileage to be driven next year. The miles driven during the past 5 years are as follows:

YEAR MILEAGE
1 3,000
2 4,000
3 3,400
4 3,800
5 3,700

a) Forecast the mileage for next year (6th year) using a 2-year moving average.
YEAR MILEAGE
1 3,000
2 4,000
3 3,400
4 3,800
5 3,700
6 3750 (3,800 + 3,700) / 2

b) Find the MAD based on the 2-year moving average. (Hint: You will have only 3 years of matched data.)

YEAR MILEAGE 2-YEAR MOVING ERROR [ERROR]

AVERAGE
1 3,000
2 4,000
3 3,400 3500 (3,000 + 4,000) /2 -100 3400-3500 100
4 3,800 3700 (4,000 + 3,400) /2 100 3800-3700 100
5 3,700 3600 (3,400 + 3,800) /2 100 3700-3600 100
MAD = 100 (100+100+100)/3

c) Use a weighted 2-year moving average with weights of .4 and .6 to forecast next year’s mileage. (The weight
of .6 is for the most recent year.) What MAD results from using this approach to forecasting? (Hint: You will
have only 3 years of matched data.)

YEAR MILEAGE 2-YEAR MOVING ERROR [ERROR] Weights

AVERAGE
1 3,000
2 4,000
3 3,400 3,600 -200 200 0.4
4 3,800 3640 160 -160 0.4
5 3,700 3640 60 -60 0.4
MAD= -6.67 0.6
1.8

d) Compute the forecast for year 6 using exponential smoothing, an initial forecast for year 1 of 3,000 miles,
and a = .5

MILEAGE FORECAST FORECAST ERROR *

NEW
YEAR
ERROR 0.5 FORECAST
1 3,000 3,000 0 0 3,000
2 4,000 3,000 1,000 500 3,500
3 3,400 3,500 -100 -50 3,450
4 3,800 3,450 350 175 3,625
5 3,700 3,625 75 38 3663
6 3663
1,325
(100+100+100)/3
The following gives the number of accidents that occurred on Florida State Highway 101 during the past 4 months:

Forecast the number of accidents that will occur in May, using least-squares regression to derive a trend equation.

MONTH # OF ACCIDENTS (y) x xy x2 ∑xy - nẍӯ

b=
January 30 1 30 1 ∑x2 - nẍ2
February 40 2 80 4
March 60 3 180 9
April 90 4 360 16
Totals 220 10 650 30
Averages 55 2.5

May 105 a= ӯ - bẍ

The regression line is y=5 + 20x, and the forecast for May (x=5) is y = 5 + 20(5) = 105
the past 4 months:

e a trend equation.

∑xy - nẍӯ 650 - 4(2.5)(55)

̿
∑x2 - nẍ2 30 - 4(2.5)^2
650 - 550
̿
30 - 25
100
̿
5
b= 20

ӯ - bẍ ̿ 55 - (20)(2.5)
a= 5
George Kyparisis owns a company that manufactures sailboats. Actual demand for George’s sailboats during each of the past f
follows:
George has forecasted that annual demand for his sailboats in year 5 will equal 5,600 sailboats. Based on this data and the mu
model, what will the demand level be for George’s sailboats in the spring of year 5?

DEMAND AVERAGE AVERAGE

YEAR YEARLY SEASONAL
SEASON DEMAND DEMANDa
1 2 3 4
Winter 1,400 1,200 1,000 900 1125 1250
Spring 1,500 1,400 1,600 1,500 1500 1250
Summer 1,000 2,100 2,000 1,900 1750 1250
Fall 600 750 650 500 625 1250
Total Average Annual Demand 5000
a
Average Seasonal Demand 1250

DEMAND
YEAR
SEASONAL INDEXb YEAR
SEASON
5 5
0.9 1260 Winter
1.2 1680 Spring
5600
1.4 1960 Summer
0.5 700 Fall

George's 5th Year sailboat Demand Level in the Spring is forecasted at 1680.
boats during each of the past four seasons was as

Based on this data and the multiplicative seasonal

SEASONAL INDEXb

0.9
1.2
1.4
0.5

ted at 1680.