Vous êtes sur la page 1sur 14

PRUEBA TRES

PROGRAMACIN DE OPERACIONES

GUERRERO TORRES MARIA CAMILA - 201324946


PADILLA ABADIA SEBASTIN - 201425019
VASQUEZ JIMENEZ MANUEL ALEJANDRO - 201428874
POINT1

WaterCraft,Inc. is a manufacturer of personal watercraft (also known as jet skis).Throughout


its first five years of operation, the Company has enjoyed a fairly steady growth in sales of
its products. The officers of the Company are preparing sales and manufacturing plans for
the coming year. A critical input to these plans involves a forecast of the level of sales that
the Company expects to achieve. Quarterly sales data for the Company during the past five
years are given.($).Forecast next year 2006.

YEAR 1 2 3 4
$ $ $ $
2001 684 584 765 892
$ $ $ $
2002 885 667 1.007 1.122
$ $ $ $
2003 1.163 993 1.313 1.545
$ $ $ $
2004 1.596 1.260 1.735 2.030
Use: Double moving average, double exponential smoothing (Holt`s Method)with
$ $ seasonal$ Effects with
$
MSE, Holt-Winter`s method for additive MSE, Holt-Winter`s
2005 2.108 1.650 2.304 2.639
method for
Double Moving average Forecasting with the model

E nT
Y E nT
Yt n t t 20 n 20 20

Forecasts for time periods 21 to 24 at time


Where
period 20:

E 2M D
Y E 1T
t t t 21 20 20
T 2(M D ) /(k 1)
Y E 2T
t t t 22 20 20
M (Y t Y t 1 Y t k 1) / k
Y E 3T
t 23 20 20
D (M t M t 1 M t k 1) / k
t
Y E 4T
24 20 20

Et is the expected base level at time


period t.
Tt is the expected trend at time period t.
Double Moving average
Double Exponential Smoothing
(Holts Method) Forecasting with the model

E nT
Y E nT
Yt n t t 20 n 20 20

Forecasts for time periods 21 to 24 at time


Where
period 20:

Et = Yt + (1-)(Et-1+ Tt-1)
Y E 1T
21 20 20
Tt = (Et Et-1) + (1-) Tt-1
Y E 2T
22 20 20
0 1 and 0 1

Y E 3T
23 20 20
Et is the expected base level at time period t.
Y E 4T
24 20 20
Tt is the expected trend at time period t.
Double Exponential Smoothing (Holts Method)
Holt-Winters Method For Additive
Seasonal Effects Forecasting with the model

Y E nT S Y 20 n E 20 nT20 S 20 n 4
tn t t tn p

Where Forecasts for time periods 21 to 24 at time

E t Yt St p (1 - )(E t 1 Tt 1 )
period 20:

T E E
t t

t 1
(1 - )T
t 1
E 1 T S
Y 21 20 20 17

E 2T S
Y

S Y E (1 - )S
t t t

t p
22 20 20

E 3 T S
18

Y 23 20 20 19

0 1 Et is the expected base level at time E 4T S


Y24 20 20 20
0 1 period t.
Tt is the expected trend at time period t.
0 1
St is the seasonal factor at time period t.
Holt-Winters Method For Additive Seasonal Effects
Holt-Winters Method For
Multiplicative Seasonal Effects Forecasting with the model

20 n E 20 nT20 S 20 n 4
E nT S
Y
Y
t n t t t n p

Where Forecasts for time periods 21 to 24 at time


period 20:
E t Yt / St p (1 - )(E t 1 Tt 1 )

( E 1T ) S
Y21 20 20 17
T E E (1 - )T
t t t 1 t 1 ( E 2T ) S
Y
S t Yt / E t (1 - )St p
22 20 20 18

(E 3T ) S
Y23 20 20 19

0 1 Et is the expected base level at time ( E 4T ) S


Y24 20 20 20
0 1 period t.
Tt is the expected trend at time period t.
0 1
St is the seasonal factor at time period t.
Holt-Winters Method For Multiplicative Seasonal
Effects
The Linear Trend Model The Quadratic Trend Model
b b X
Y b b X b X
Y
t 0 1 1t t 0 1 1t 2 2t

where X 1t t where X 1t t and X 2t t 2

Forecasts for time periods 21 to 24 at Forecasts for time periods 21 to 24 at


time period 20: time period 20:

b b X
Y b b X b X
Y
21 0 1 121 21 0 1 121 2 2 21

b b X
Y b b X b X
Y
22 0 1 122 22 0 1 122 2 2 22

b b X
Y b b X b X
Y
23 0 1 123 23 0 1 123 2 2 23

b b X
Y b b X b X
Y
24 0 1 124 24 0 1 124 2 2 24
The Linear Trend Model
The Quadratic Trend Model
Measuring Accuracy: The mean square error

MSE =
n
Yi Y
i 2

i 1 n

Vous aimerez peut-être aussi