Operation Mangement Chapter 3

Operations Management: Processes and
Supply Chains
Twelfth Edition, Global Edition
Chapter 8
Forecasting
Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

What is a Forecast?
• Forecast
– A prediction of future events used for planning
purposes.

Demand Patterns (1 of 5)
• Time series
– The repeated observations of demand for a service or
product in their order of occurrence
• There are five basic time series patterns
1. Horizontal
2. Trend
3. Seasonal
4. Cyclical
5. Random

Figure 8.1 Patterns of Demand
(a) Horizontal: Data cluster about a horizontal line

Figure 8.1 [continued]
(b) Trend: Data consistently increase or decrease

(c) Seasonal: Data consistently show peaks and valleys

(d) Cyclical: Data reveal gradual increases and decreases over
extended periods

Demand Management Options (1 of 2)
• Demand Management
– The process of changing demand patterns using one
or more demand options

Demand Management Options (2 of 2)
• Complementary Products
• Promotional Pricing
• Prescheduled Appointments
• Reservations
• Revenue Management
• Backlogs
• Backorders and Stockouts

Key Decisions on Making Forecasts
• Deciding What to Forecast

– Level of aggregation
– Units of measurement
• Choosing the Type of Forecasting Technique
– Judgment methods
– Causal methods
– Time-series analysis
– Trend projection using regression

Forecast Error
• For any forecasting method, it is important to measure the

accuracy of its forecasts.
• Forecast error is simply the difference found by subtracting
the forecast from actual demand for a given period, or
Et = Dt − Ft
where
Et = forecast error for period t
Dt = actual demand in period t
Ft = forecast for period t

Measures of Forecast Error (1 of 3)
Cumulative sum of Standard deviation

forecast errors (Bias)
( )
2
 Et − E
CFE =  Et  =
n −1
Average forecast error Mean Absolute Deviation
E=
CFE  Et
MAD =
n n
Mean Squared Error Mean Absolute Percent Error

 Et
2
MAPE =
( E t Dt ) (100 )
MSE =
n n

Figure 8.2(B) Detailed Calculations of Forecast Errors
Blank Actual Forecast Error Absolute
Error Error Error
Errorcap
^2 2 Pct Error
Absolute Pct
Error
Past period 1 39 41 −2 2 4 5.128%

Past period 2 37 43 −6 6 36 16.216%
Past period 3 55 45 10 10 100 18.182%
Past period 4 40 50 −10 10 100 25%
Past period 5 59 51 8 8 64 13.559%
Past period 6 63 56 7 7 49 11.111%
Past period 7 41 61 −20 20 400 48.78%
Past period 8 57 60 −3 3 9 5.236%
Past period 9 56 62 −6 6 36 10.714%
Past period 10 54 63 −9 9 81 16.667%
Totals 501 Blank −31 81 879 170.621%
Average 50.1 Blank −3.1 8.1 87.9 17.062%
Next period forecast Blank 0 (Bias) (MAD) (MSE) (MAPE)
Blank Blank Blank Blank std err 9.883 Blank

Figure 8.2(C) Error Measures
Measure Value
Error Measures Blank
CFC (Cumulative Forecast Error) −31
MAD (Mean Absolute Deviation) 8.1
MSE (Mean Squared Error) 87.9
Standard Deviation of Errors 9.883
MAPE (Mean Absolute Percent Error) 17.062%

Example 1 (1 of 4)
The following table shows the actual sales of upholstered chairs for a
furniture manufacturer and the forecasts made for each of the last eight
months.
Calculate CFE, MSE, σ, MAD, and MAPE for this product.

Example 1 (2 of 4)
Using the formulas for the measures, we get:
Cumulative forecast error (mean bias)
CFE = −15
Average forecast error (mean bias):
CFE 15
E= = = −1.875
n 8
Mean squared error:
 Et
2
5,275
MSE = = = 659.4
n 8

Example 1 (3 of 4)
Standard deviation:
 Et − ( −1.875)
2
 = = 27.4
n −1
Mean absolute deviation:

 Et 195
MAD = = = 24.4
n 8
Mean absolute percent error:
MAPE =
( E t Dt ) (100 )
=
81.3%
= 10.2%
n 8

Example 1 (4 of 4)
• A CFE of −15 indicates that the forecast has a slight bias to
overestimate demand.
• The MSE, σ, and MAD statistics provide measures of forecast error
variability.
• A MAD of 24.4 means that the average forecast error was 24.4 units
in absolute value.
• The value of σ, 27.4, indicates that the sample distribution of forecast
errors has a standard deviation of 27.4 units.
• A MAPE of 10.2 percent implies that, on average, the forecast error
was about 10 percent of actual demand.
These measures become more reliable as the number of periods of
data increases.
Judgment Methods
• Other methods (casual, time-series, and trend projection

using regression) require an adequate history file, which
might not be available.
• Judgmental forecasts use contextual knowledge gained
through experience.
– Salesforce estimates
– Executive opinion
– Market research
– Delphi method

Causal Methods: Linear Regression
• Dependent variable – The variable that one wants to forecast
• Independent variable – The variable that is assumed to affect
the dependent variable and thereby “cause” the results
observed in the past
• Simple linear regression model is a straight line
Y = a + bX
where
Y = dependent variable
X = independent variable
a = Y-intercept of the line
b = slope of the line
Linear Regression (1 of 2)
Figure 8.3 Linear Regression Line Relative to Actual Demand

Linear Regression (2 of 2)
• The sample correlation coefficient, r
– Measures the direction and strength of the relationship between
the independent variable and the dependent variable.
– The value of r can range from −1.00  r  1.00
• The sample coefficient of determination, r 2
– Measures the amount of variation in the dependent variable
about its mean that is explained by the regression line
– The values of r 2 range from0.00  r ²  1.00
• The standard error of the estimate, syx
– Measures how closely the data on the dependent variable
cluster around the regression line

Example 2 (1 of 4)
The supply chain manager seeks a better way to forecast the demand
for door hinges and believes that the demand is related to advertising
expenditures. The following are sales and advertising data for the past
5 months:
Month Sales (thousands of units) Advertising (thousands of $)
1 264 2.5
2 116 1.3
3 165 1.4
4 101 1.0
5 209 2.0
The company will spend $1,750 next month on advertising for the
product. Use linear regression to develop an equation and a forecast
for this product.

Example 2 (2 of 4)
We used POM for Windows to determine the best values

of a, b, the correlation coefficient, the coefficient of
determination, and the standard error of the estimate
a = −8.135
b = 109.229
r = 0.980
r 2 = 0.960
syx = 15.603
The regression equation is Y = −8.135 + 109.229X

Example 2 (3 of 4)
The r of 0.98 suggests an unusually strong positive relationship
between sales and advertising expenditures. The coefficient of
determination, r 2 , implies that 96 percent of the variation in sales
is explained by advertising expenditures.
Figure 8.4 Linear

Regression Line
for the Sales and
Advertising Data
Using POM for
Windows

Example 2 (4 of 4)
• Forecast for month 6:

Y = −8.135 + 109.229X
Y = −8.135 + 109.229 (1.75 )
Y = 183.016 or 183,016 units

Time Series Methods
• Naïve forecast
– The forecast for the next period equals the demand
for the current period (Forecast = Dt)
• Horizontal Patterns: Estimating the average
– Simple moving average
– Weighted moving average
– Exponential smoothing

Simple Moving Averages
• Specifically, the forecast for period t + 1 can be

calculated at the end of period t (after the actual demand
for period t is known) as
Sum of last n demands Dt + Dt −1 + Dt −2 +    + Dt −n+1

Ft +1 = =
n n
where
Dt = actual demand in period t
n = total number of periods in the average
Ft+1 = forecast for period t + 1

Example 3 (1 of 2)
a. Compute a three-week moving average forecast for the
arrival of medical clinic patients in week 4. The numbers
of arrivals for the past three weeks were as follows:
Week Patient Arrivals
1 400
2 380
3 411
b. If the actual number of patient arrivals in week 4 is 415,

what is the forecast error for week 4?
c. What is the forecast for week 5?
Example 3 (2 of 2)
a. The moving average forecast at Week Patient Arrivals

1 400
the end of week 3 is:
2 380
411 + 380 + 400 3 411
F4 = = 397.0
3
b. The forecast error for week 4 is
E4 = D4 − F4 = 415 − 397 = 18
c. The forecast for week 5 requires the actual arrivals from

weeks 2 through 4, the three most recent weeks of data
415 + 411 + 380
F5 = = 402.0
3
Weighted Moving Averages
In the weighted moving average method, each historical

demand in the average can have its own weight, provided
that the sum of the weights equals 1.0.
The average is obtained by multiplying the weight of each
period by the actual demand for that period, and then
adding the products together
Ft +1 = W1D1 + W2D2 +    + Wn Dt − n +1

Exponential Smoothing (1 of 2)
• A sophisticated weighted moving average that calculates

the average of a time series by implicitly giving recent
demands more weight than earlier demands
• Requires only three items of data
– The last period’s forecast
– The actual demand for this period
– A smoothing parameter, alpha (α), where 0  α  1.0
• The equation for the forecast is
Ft +1 = α (Demand this period ) + (1 −  )(Forecast calculated last period)
= αDt + (1 −  )Ft

Exponential Smoothing (2 of 2)
• The emphasis given to the most recent demand levels

can be adjusted by changing the smoothing parameter.
• Larger α values emphasize recent levels of demand and
result in forecasts more responsive to changes in the
underlying average.
• Smaller α values are analogous to increasing the value of
n in the moving average method and giving greater
weight to past demand.

Example 4 (1 of 3)
a. Reconsider the patient arrival data in Example 14.3. It is

now the end of week 3 so the actual arrivals is known to
be 411 patients. Using α = 0.10, calculate the
exponential smoothing forecast for week 4.
b. What was the forecast error for week 4 if the actual
demand turned out to be 415?
c. What is the forecast for week 5?

Example 4 (2 of 3)
a. To obtain the forecast for week 4, using exponential

smoothing with and the initial forecast of 390*, we
calculate the forecast for week 4 as:
F 4 = 0.10 ( 411) + 0.90 ( 390 ) = 392.1
Thus, the forecast for week 4 would be 392 patients.
* POM for Windows and OM Explorer simply use the actual
demand for the first week as the default setting for the
initial forecast for period 1, and do not begin tracking
forecast errors until the second period.

Example 4 (3 of 3)

E4 = 415 − 392 = 23
c. The new forecast for week 5 would be
F5 = 0.10 ( 415 ) + 0.90 ( 392.1) = 394.4 or 394 patients.

Trend Patterns: Using Regression
• A trend in a time series is a systematic increase or

decrease in the average of the series over time
• Trend Projection with Regression accounts for the trend
with simple regression analysis.
• The regression equation is Ft = a + bt

Seasonal Patterns: Using Seasonal Factors
• Multiplicative seasonal method

– A method whereby seasonal factors are multiplied by
an estimate of average demand to arrive at a
seasonal forecast.
• Additive seasonal method
– A method in which seasonal forecasts are generated
by adding a constant to the estimate of average
demand per season.

Multiplicative Seasonal Method
Multiplicative seasonal method

1. For each year, calculate the average demand for each
season by dividing annual demand by the number of
seasons per year.
2. For each year, divide the actual demand for each
season by the average demand per season, resulting in
a seasonal factor for each season.
3. Calculate the average seasonal factor for each season
using the results from Step 2.
4. Calculate each season’s forecast for next year.
Example 6 (1 of 5)
The manager of the Stanley Steemer carpet cleaning

company needs a quarterly forecast of the number of
customers expected next year. The carpet cleaning business
is seasonal, with a peak in the third quarter and a trough in
the first quarter.
The manager wants to forecast customer demand for each
quarter of year 5, based on an estimate of total year 5
demand of 2,600 customers.
The table on the following slides shows the quarterly
demand data from the past 4 years along with the calculation
of the seasonal factor for each week.
Example 6 (2 of 5)

Example 6 (3 of 5)

Example 6 (4 of 5)
Average Seasonal Factor Quarterly Forecasts

Average Seasonal Quarter Forecast
Quarter Factor 650 
times 0.2043==132.795
132.795
1 650 0.2043
1 0.2043
2 650times
650 1.2979 == 843.635
1.2979 843.635
2 1.2979
3 650times
650 2.001 = = 1,300.065
2.001 1,300.065
3 2.0001
4 650times
650 0.4977== 323.505
0.4977 323.505
4 0.4977

Example 6 (5 of 5)
Figure 8.7 Demand Forecasts Using the Seasonal Forecasting
Solver of OM Explorer

Criteria for Selecting Time-Series Method
• Criteria
1. Minimizing bias (CFE)
2. Minimizing MAPE, MAD, or MSE
3. Maximizing r 2 for trend projections using regression
4. Using a holdout sample analysis
5. Using a tracking signal
6. Meeting managerial expectations of changes in the
components of demand
7. Minimizing the forecast errors in recent periods

Choosing a Time-Series Method (1 of 2)
• Using Statistical Criteria

1. For projections of more stable demand patterns, use
lower α values or larger n values to emphasize
historical experience.
2. For projections of more dynamic demand patters, use
higher α values or smaller n values.

Choosing a Time-Series Method (2 of 2)
• Holdout sample
– Actual demands from the more recent time periods in
the time series that are set aside to test different
models developed from the earlier time periods.

Tracking Signals (1 of 3)
• A tracking signal is a measure that indicates whether a
method of forecasting is accurately predicting actual
changes in demand.
CFE CFE
Tracking signal = or
MAD MADt
Each period, the CFE and MAD are updated to reflect

current error, and the tracking signal is compared to some
predetermined limits.

• The MAD can be calculated as the simple average of all

absolute errors or as a weighted average determined by
the exponential smoothing method
MADt =  | Et | + (1 −  )MADt −1
If forecast errors are normally distributed with a mean of 0,

the relationship between σ and MAD is simple
  
 =   (MAD )  1.25 (MAD )
 2
MAD = 0.7978  0.8 where  = 3.1416

Figure 8.8 Tracking Signal

Insights Into Effective Demand Forecasting
• Big Data
– Data sets that are so large or complex that traditional
data processing applications are inadequate to deal
with them.
• Big Data is characterized by:
– Volume
– Variety
– Velocity

Forecasting as a Process

Using Multiple Forecasting Methods
• Combination forecasts
– Forecasts that are produced by averaging
independent forecasts based on different methods,
different sources, or different data
• Focus forecasting
– A method of forecasting that selects the best forecast
from a group of forecasts generated by individual
techniques.

Forecasting Principles
Table 8.2 Some Principles for the Forecasting Process
• Better processes yield better forecasts.

• Demand forecasting is being done in virtually every company, either formally or
informally. The challenge is to do it well—better than the competition.
• Better forecasts result in better customer service and lower costs, as well as better
relationships with suppliers and customers.
• The forecast can and must make sense based on the big picture, economic outlook,
market share, and so on.
• The best way to improve forecast accuracy is to focus on reducing forecast error.
• Bias is the worst kind of forecast error; strive for zero bias.
• Whenever possible, forecast at more aggregate levels. Forecast in detail only where
necessary.
• Far more can be gained by people collaborating and communicating well than by using
the most advanced forecasting technique or model.
Source: From Thomas F. Wallace and Robert A. Stahl, Sales Forecasting: A New Approach (Cincinnati, O H: T. E. Wallace
& Company, 2002), p. 112. Copyright © 2002 T.E. Wallace & Company. Used with permission.

Adding Collaboration to the Process
CPFR Collaborative Planning, Forecasting, and

Replenishment
• A process for supply chain integration that allows a
supplier and its customers to collaborate on making the
forecast by using the Internet.

Solved Problem 1 (1 of 2)
Chicken Palace periodically offers carryout five-piece chicken dinners
at special prices. Let Y be the number of dinners sold and X be the
price. Based on the historical observations and calculations in the
following table, determine the regression equation, correlation
coefficient, and coefficient of determination. How many dinners can
Chicken Palace expect to sell at $3.00 each?
Observation Price (X) Dinners Sold (Y)
1 $2.70 760
2 $3.50 510
3 $2.00 980
4 $4.20 250
5 $3.10 320
6 $4.05 480
Total $19.55 3,300
Average $3.26 550

We use the computer to calculate the best values of a, b, the
correlation coefficient, and the coefficient of determination
a = 1,454.60
b = −277.63
r = −0 .84
r ² = 0.71
The regression line is
Y = a + bX = 1,454.60 − 277.63X
For an estimated sales price of $3.00 per dinner
Y = a + bX = 1,454.60 − 277.63 ( 3.00 )
= 621.71 or 622dinners
The Polish General’s Pizza Parlor is a small restaurant catering to
patrons with a taste for European pizza. One of its specialties is Polish
Prize pizza. The manager must forecast weekly demand for these special
pizzas so that he can order pizza shells weekly. Recently, demand has
been as follows:
Week Pizzas Week Pizzas
June 2 50 June 23 56
June 9 65 June 30 55
June 16 52 July 7 60
a. Forecast the demand for pizza for June 23 to July 14 by using the
simple moving average method with n = 3 then using the weighted
moving average method with weights of 0.50, 0.30, and 0.20, with .50
applying to the most recent demand.
b. Calculate the MAD for each method.
a. The simple moving average method and the weighted

moving average method give the following results:
Current Simple Moving Average Weighted Moving Average Forecast for Next Week
Week Forecast for Next Week
52 plus 65 plus 50, over 3 equals left bracket left paranthesis 0.5 times 52 right paranthesis + left
June 16 52 or
55.7 + 65
56. + 50
= 55.7 or 56 (
 0.5  52
parantesis ) ( ) (
+ 0.365 right
0.3 times 65 +paranthesis )
0.2  50 + left
= 55.5or
paranthesis
56 0.2
3 times 50 right paranthesis right bracket = 55.5 or 56
56 plus 52 plus 65, over 3 equals left bracket left paranthesis 0.5 times 56 right paranthesis + left
June 23 56 or
57.7 + 52
58. + 65
= 57.7 or 58 (
 0.5  56
parantesis ) (
0.3 times ) (
+ 0.352 right )
0.2  65 + left
52 +paranthesis = 56.6or 57 0.2
paranthesis
55 plus 56 plus 52, over 3 equals
June 30
55 + 56 + 52
54.3 or 54. = 54.3 or 54 (
parantesis ) ( ) ( )
left bracket left paranthesis 0.5 times 55 right paranthesis + left
 0.5  55 + 0.3  56 + 0.2  52  = 54.7or 55
0.3 times 56 right paranthesis +left paranthesis 0.2
60 plus 55 plus 56, over 3 equals
July 7
60 + 55 + 56
57.0 or 57. = 57.0 or 57 (
parantesis
) ( ) ( )
left bracket left paranthesis 0.5 times 60 right paranthesis + left
 0.5  60 + 0.3  55 + 0.2  56  = 57.7or 58
0.3 times 55 right paranthesis + left paranthesis 0.2

b. The mean absolute deviation is calculated as follows:
For this limited set of data, the weighted moving average method resulted
in a slightly lower mean absolute deviation. However, final conclusions
can be made only after analyzing much more data.

The monthly demand for units manufactured by the Acme Rocket Company has
been as follows:
Month Units Month Units
May 100 September 105
June 80 October 110
July 110 November 125
August 115 December 120
a. Use the exponential smoothing method to forecast June to January. The

initial forecast for May was 105 units; α = 0.2.
b. Calculate the absolute percentage error for each month from June through
December and the MAD and MAPE of forecast error as of the end of
December.
c. Calculate the tracking signal as of the end of December. What can you say
about the performance of your forecasting method?
a.
Current Month, Calculating Forecast for Next Month Forecast for
t Ft +1 =  Dt + (1 −  )Ft Month t + 1
May 0.20.2 (100

times ) ++ 0.8
100 (105105
0.8times ) = =104.0
104.0or
or 104
104 June
June 0.20.2 ( 8080

times ) ++ 0.8 (104.0
0.8times ) = =99.2
104.0 99.2 or 99
or 99 July
July 0.20.2 (110

times ) ++ 0.8
110 ( 99.299.2
0.8times ) = =101.4
101.4 or
or 101
101 August
August 0.20.2 (115115

times ) ++ 0.8 (101.4
0.8times ) == 104.1
101.4 or104
104.1 or 104 September
September 0.20.2 (105105

times ) ++ 0.8 (104.1
0.8times ) = =104.3
104.1 or104
104.3 or 104 October
October 0.20.2 (110110

times ) ++ 0.8 (104.3
0.8times ) ==105.4
104.3 or105
105.4 or 105 November
November 0.20.2 (125125

times ) ++ 0.8 (105.4
0.8times ) = =109.3
105.4 109.3or
or 109
109 December
December 0.2
0.2 (120120
times ) ++ 0.8 (109.3109.3
0.8times ) = 111.4
= 111.4or
or111
111 January

b.
MAD =
 Et
=
87
= 12.4 MAPE =
( E t Dt ) (100 )
=
83.7%
= 11.96%
n 7 n 7
c. As of the end of December, the cumulative sum of forecast
errors (C F E) is 39. Using the mean absolute deviation
calculated in part (b), we calculate the tracking signal:
CFE 39
Tracking signal = = = 3.14
MAD 12.4
The probability that a tracking signal value of 3.14 could be

generated completely by chance is small. Consequently, we
should revise our approach. The long string of forecasts lower
than actual demand suggests use of a trend method.

The Northville Post Office experiences a seasonal pattern of daily mail volume
every week. The following data for two representative weeks are expressed in
thousands of pieces of mail:
Day Week 1 Week 2

Sunday 5 8
Monday 20 15
Tuesday 30 32
Wednesday 35 30
Thursday 49 45
Friday 70 70
Saturday 15 10
Total 224 210
a. Calculate a seasonal factor for each day of the week.
b. If the postmaster estimates 230,000 pieces of mail to be sorted next week,

forecast the volume for each day.


b. The average daily mail volume (in pieces of mail) is expected
230,000
to be = 32,857
7
Using the average seasonal factors calculated in part (a),

we obtain the following forecasts:

SUPLEMEN

Taksonomi Peramalan Berdasarkan
Derajat Kuantifikasi (1)
MODEL
MODEL
KUALITATIF
KUALITATIF
REGRESI
REGRESI
PERAMALAN
PERAMALAN TIME SERIES
TIME SERIES RATA-RATA
RATA-RATA
MOVING
SMOOTHING
SMOOTHIN MOVING AVERAGE
AVERAGE
G
MODEL
MODEL
KUANTITATIF
KUALITATIF
EXPONENTIAL
SMOOTHING
KAUSAL
CAUSAL
Copyright © 2019 Pearson Education, Ltd. All Rights Reserved69

JANGKA WAKTU PERAMALAN
• Jangka Panjang : > 1 tahun
• Jangka Menengah : 3 bulan – 1 tahun
• Jangka Pendek: 0 – 3 bulan

JANGKA PENDEK (0-3 bulan)
• Untuk meramalkan permintaan akan suatu produk atau jasa
• Umumnya digunakan analisa time series, causal dan judgment
• Untuk jangka pendek manajer jarang ingin menunggu.
• Judgment digunakan apabila data masa lalu tidak tersedia
• Keputusan untuk manajemen persediaan, penjadwalan final
assembly, penjadwalan tenaga kerja, master production scheduling
(MPS)
• Kritis untuk day-to-day planning
• Umumnya diukur dalam hari atau minggu

JANGKA MENENGAH (3 bln-1 thn)
• produksi, MPS, pembelian dan distribusi

• Umum diukur dalam minggu atau bulan

JANGKA PANJANG (> 1 thn)
• Peramalan biasanya dikembangkan untuk menghitung

penjualan total
• 3 (tiga) jenis keputusan: lokasi fasilitas, perencanaan
kapasitas dan pemilihan proses
• Umumnya digunakan metoda causal dan judgment

Example 3
a. The moving average forecast at Week Patient Arrivals

1 400
the end of week 3 is:
2 380
411 + 380 + 400 3 411
F4 = = 397.0
3
E4 = D4 − F4 = 415 − 397 = 18
c. The forecast for week 5 requires the actual arrivals from

weeks 2 through 4, the three most recent weeks of data
415 + 411 + 380
F5 = = 402.0
3
75
Simple Moving Average (SMA)

Minggu Demand SMA-3 SMA-6
1 650 F4=(650+678+720)/3
2 678 = 682,67
Hitung tingkat
3 720 akurasi metode
4 785 682,67 F7=(650+678+
5 859 727,67 720+785+859+920)/6
= 768,67
6 920 788,00
7 850 854,67 768,67
8 758 876,33 802,00
9 892 842,67 815,33
10 920 833,33 844,00
11 789 856,67 866,5
12 844 867,00 854,83
SMA Plotting
1000
800
Demand
Demand
600
SMA-3
400 SMA-6
200
0
1 2 3 4 5 6 7 8 9 10 11 12
Minggu

• Jadi peramalan untuk minggu ke 13 adalah sbb :
- SMA-3 : (920+789+844)/3 = 851
- SMA-6 : (850+758+892+920+789+844)/6 = 842,17
• Jika pada minggu ke 13 ternyata demand yang terjadi

adalah 849, berapakah peramalan untuk periode ke 14 ?
- SMA-3 : (789+844+849)/3 = 827,33
- SMA-6 : (758+892+920+789+844+849)/6 = 842

Double Moving Average (DMA)
• Digunakan untuk pola data trend
• Merupakan SMA yang dirata-ratakan kembali untuk
mendapatkan trend.
• SMA digunakan pada waktu t (S’t)
• Penyesuaian merupakan perbedaan antara SMA dan
DMA pada waktu t (S’t-S’’t)
• Penyesuaian digunakan untuk trend dari periode t ke
periode t+1 (atau periode t+m jika diramalkan untuk m
periode mendatang)

79

Dt + Dt −1 + Dt −2 + ... + Dt − n +1
S 't =
n
Formula DMA
S 't + S 't −1+ S 't − 2 +... + S 't − n +1
S ' 't =
n
at = S 't +( S 't − S ' 't ) = 2S 't − S ' 't
2
bt = ( St' − St'' )
N −1
Ft + m = at + bt m

80

Hitung tingkat akurasi
metode
Periode Demand SMA-4 DMA-4 a b Ft

1 140 S’t=(140+159+136+157)/4 F8=165,75+4,17*1
2 159
3 136 S’’t=(148+156,25+149,25+159,5)/4
4 157 148,00
5 173 156,25 a = (2*159,5-153,25)
b = (2/3*(159,5-153,25))
6 131 149,25
7 177 159,50 153,25 165,75 4,17
8 188 167,25 158,06 176,44 6,13 169,92
9 154 162,50 159,63 165,38 1,92 182,56
10 179 174,50 165,94 183,06 5,71 167,29
11 188,77

81
Simple Exponential Smoothing (SES) Hitung tingkat

akurasi metode
Minggu Demand SES 0,1 SES 0,6
1 820 820 820
2 775 820 820
3 680 815.5 793
4 655 801.95 725.2
5 750 787.255 683.08
6 802 783.5295 723.232
7 798 785.3766 770.4928
8 689 786.6389 786.9971
9 775 776.875 728.1988
10 776.6875 756.2795

Simple Exponential Smoothing (SES)
Sim ple Exponential Sm oothing Plotting
1000
800
Demand
Demand
600
SES 0,1
400
SES 0,6
200
0
1 2 3 4 5 6 7 8 9
Minggu

Persamaan SES dan SMA
• Kedua memiliki asumsi bahwa demand tetap (stationary)
• Kedua metoda tergantung pada satu parameter. Jika SMA adalah N, jumlah perioda
dalam SMA, sedangkan SES parameternya α , kontanta pemulusan (smoothing
constant)
• Ketika α = 2/(N+1), maka kedua metoda akan memiliki distribusi kesalahan peramalan
yang sama. Ini berarti kedua nya akan memberikan tingkat akurasi yang sama, tetapi
tidak memberikan hasil peramalan yang sama.

Perbedaan SES dan SMA
• SES memberikan bobot yang sama untuk semua data masa lalu, SMA hanya akan
memberikan bobot yang sama untuk N perioda yang terakhir
• Pada SMA dibutuhkan semua data N masa lalu yang terakhir, sedangkan SES hanya
membutuhkan satu data masa lalu yang terakhir. Hal ini merupakan kelebihan dari
SES dibandingkan dengan SMA

Double Exponential Smoothing (DES) Satu
Parameter
• Disebut juga Metode Linear Brown
• Untuk pola data trend
• Hampir sama dengan Double Moving Average
• Penyesuaian dari SES dilakukan dengan
penambahan satu parameter

86
Double Exponential Smoothing (DES) Satu

Parameter
Rumus :
St' = Dt +(1 −  ) St' −1
St'' = St' + (1 −  ) St''−1

at = S 't +( S 't − S ' 't ) = 2S 't − S ' 't
 ' ''
bt = ( St − St )
1−
Ft + m = at + bt m

87
Contoh DES satu parameter α = 0.2

Minggu Dt S't S''t a b Ft
1 143.00 143.00 143.00 143.00 0.00
2 152.00 144.80 143.36 146.24 0.36 143.00
3 161.00 148.04 144.30 151.78 0.94 146.60
4 139.00 146.23 144.68 147.78 0.39 152.72
5 137.00 144.39 144.62 144.15 -0.06 148.17 Hitung
6 174.00 150.31 145.76 154.86 1.14 144.09 tingkat
akurasi
7 142.00 148.65 146.34 150.96 0.58 155.99
metode
8 141.00 147.12 146.49 147.74 0.16 151.53
9 162.00 150.09 147.21 152.97 0.72 147.90
10 153.69 m = 1
11 154.41 m = 2
12 155.13 m = 3

Contoh DES satu parameter α = 0.2
DES plotting
200
150
Demand
Dt
100
Ft
50
0
1 2 3 4 5 6 7 8 9 10 11 12
Minggu

JANGKA MENENGAH
n
• Konstan ^ ^ ^ d t
t =1
D (t ) = a a= n
^ ^ ^
• Regresi
D (t ) =linier
a + bt Regresi Kuadratis
Dt = a + bt + ct 2
N N N N
N  t .d (t ) −  d (t ) t  d (t ) n n n n
 t.dt  dt t − N  t .dt

^ ^ 2 2
t =1 t =1 t =1 t =1
b = 2 a= N
N N b= t =1
c = t =1 t =1 t =1
N t − (  t )
2 n n n
t  (t 2) − n t
2 2 4
t =1
t =1
t =1 t =1 t =1
n n
 dt − C  t
2
^
2t 2t
D (t ) = a + u cos + v sin a= t =1 t =1
N N N
• Siklis

Contoh
• Dari data 12 bulan terakhir tercatat penjualan produk X:
t 1 2 3 4 5 6 7 8 9 10 11 12
dt 140 159 136 157 173 181 177 188 154 179 180 160
Gambar
diagram
Pencar:

Metoda Konstan
t Dt Ft' e = Dt - Ft' e2 SEE
1 140 165.33 -25.33 641.78
2 159 165.33 -6.33 40.11
3 136 165.33 -29.33 860.44
4 157 165.33 -8.33 69.44
5 173 165.33 7.67 58.78
6 181 165.33 15.67 245.44
7 177 165.33 11.67 136.11
8 188 165.33 22.67 513.78
9 154 165.33 -11.33 128.44
10 179 165.33 13.67 186.78
11 180 165.33 14.67 215.11
12 160 165.33 -5.33 28.44
1984 3,124.67 16.85

Metoda Konstan
n
d t
t =1 '
a= d t
N

Metoda Linier
t Dt t.Dt t2 Ft' = 149 + 3t e = Dt - Ft' e2
1 140 140 1 152 -12 144
2 159 318 4 155 4 16
3 136 408 9 158 -22 484
4 157 628 16 161 -4 16
5 173 865 25 164 9 81
6 181 1,086 36 167 14 196
7 177 1,239 49 170 7 49
8 188 1,504 64 173 15 225
9 154 1,386 81 176 -22 484
10 179 1,790 100 179 0 0
11 180 1,980 121 182 -2 4
12 160 1,920 144 185 -25 625
78 1984 13,264 650 2,022 -38 2,324

Metoda Linier
n n n
N  t .dt −  dt  t
12 13264 − 1984 78
t =1 t =1 t =1
b= 2
𝑏=
12 650 − (78)2
n n
N t − (  t )
2
159168 −154752 4416
𝑏= =
t =1
t =1 7800 −6084 1716
𝑏 = 2.57 ≈ 3
n n
 dt bt 𝑎 = 𝐷𝑡 − 𝑏. 𝑡ҧ
t =1 t =1
a= − = dt − b.t
N N 𝑎 = 165.33 − 2.57 6.5 = 148.61 ≈ 149
𝐹𝑡 ′ = 149 + 3𝑡

Metoda Kuadratis
t dt t.dt t2 t2.dt t4 dt' e=dt- e2

dt'
-6 140 -840 36 5040 1296 118.36 21.64 468.29
-5 159 -795 25 3975 625 127.82 31.18 972.19
-4 136 -544 16 2176 256 143.88 -7.88 62.09
-3 157 -471 9 1413 81 146.74 10.26 105.27
-2 173 -346 4 692 16 147.6 23.4 547.56
-1 181 -181 1 181 1 152.46 28.54 814.53
1 177 177 1 171 1 156.86 20.14 405.62
2 188 376 4 752 16 158.4 29.6 876.16
3 154 462 9 1386 81 159.96 -5.96 35.52
4 179 716 16 2864 1256 161.48 17.52 306.95
5 180 900 25 4500 625 149.82 30.18 910.83
6 160 960 36 5760 1296 144.76 15.24 232.26
 1984 404 182 28710 4550 5737.2
7
Metoda Kuadratis
404
b= = 2.22  2.2
n n n n 182
 t.dt  dt  t − N  t 2.dt
2
t =1 t =1 t =1
b= n
c = t =1
n n
2
t2  (t 2) − n  t 4 (1984 )(182 ) − 12(28910 ) 361088 − 346920
t =1 t =1 t =1 c= =
(182)
2
− 12(4550 ) 33124 − 54600
n n
 dt − C  t 2 14168
a= t =1 t =1 = = −0.6597  0.66
N − 21476
5737. 27
SEE = 1984 − 0.66(182 ) 1984 − 120 .12
12 − 3 a= =
12 12
5737. 27
= = 155 .32
9
= 25. 25  25
 dt' = 155 .32 + 2.2t + 0.66t 2
Pemilihan Metoda Terbaik & Hasil
Peramalan
• Metode yang dipilih adalah metode peramalan linier
Konstan Linier Kuadratis

SEE 17 15 25
• Dt' = 149 + 3t
t 13 14 15 16 17 18 19 20 21 22 23 24
Dt' 188 191 194 197 200 203 206 209 212 215 218 221

98
Tracking Signal (1)

Tracking Signal adalah sebuah ukuran yang mengidentifikasi apakah sebuah metode
peramalan dapat secara akurat memprediksi perubahan permintaan aktual.
Tracking signal = CFE/MAD
CFE =  Et
Setiap periode, CFE dan MAD di update untuk memperlihatkan error terbaru dan tracking
signal dibandingkan terhadap batasan yang sudah ditentukan

99
Tracking Signal (2)

MAD dapat dihitung dengan dua cara
1. Sebagai simple average dari semua absolute error
 Et
MAD =
n
2. Sebagai weighted average pada metode exponential

smoothing
MAD =  Et + (1 −  ) MADt − 1

Tracking Signal (3)
Model kedua mempunyai kelebihan yaitu jumlah data lebih
sedikit
Nilai α tidak harus sama dengan yang digunakan dalam

peramalan tetapi biasanya digunakan α kecil sekitar 0,1
untuk mengurangi efek error masa lalu.
100
101
Tracking Signal (4)

• Jika peramalan berdistribusi normal dengan rata-rata 0,
hubungan antara σ dan MAD adalah sbb :
 = (  / 2 )(MAD)  1.25( MAD)
MAD = 0.7978  0.8
• Hubungan di atas memungkinkan penggunaan tabel

distribusi normal untuk menentukan batas tracking signal.
• Pemilihan batas tracking signal melibatkan trade-off
antara biaya peramalan dan biaya pemeriksaan
permasalahan

102
Tracking Signal
CFE
• TS = ,CFE = Cumulative Forecast Error
MAD
• Beberapa buku menuliskan CFE sebagai RSFE
RSFE = Running Sum of Forecast Error, sehingga dapat

dituliskan sebagai
RSFE
TS =
MAD

Tracking Signal (5)
Control Limit Equivalent P e r c e n t o f A r e a
Spread Number of σ W i t h i n C o n t r o l
(number of MAD) L i m i t s
± 1 . 0 ± 0 . 8 0 5 7 . 6 2
± 1 . 5 ± 1 . 2 0 7 6 . 9 8
± 2 . 0 ± 1 . 6 0 8 9 . 0 4
± 2 . 5 ± 2 . 0 0 9 5 . 4 4
± 3 . 0 ± 2 . 4 0 9 8 . 3 6
± 3 . 5 ± 2 . 8 0 9 9 . 4 8
± 4 . 0 ± 3 . 2 0 9 9 . 8 6
103
104
Tracking Signal (SES , α = 0,6)
Actual Forecast error CFE |error| Cum |e| MAD TS

1 820 820 0 0 0 0 0 0
2 775 820 -45 -45 45 45 22.5 -2
3 680 793 -113 -158 113 158 52.66667 -3
4 655 725.2 -70.2 -228.2 70.2 228.2 57.05 -4
5 750 683.08 66.92 -161.28 66.92 295.12 59.0240 -2.7324
6 802 723.232 78.7680 -82.512 78.768 373.888 62.3147 -1.3241
7 798 770.4928 27.5072 -55.0048 27.5072 401.3952 57.3422 -0.9592
8 689 786.9971 -97.9971 -153.002 97.9971 499.3923 62.4240 -2.4510
9 775 728.1988 46.8012 -106.201 46.8012 546.1935 60.6882 -1.7499
10 756.2795

105
Tracking Signal (SES , α = 0,6)

0 0
1 2 3 4 5 6 7 8 9
-0,5
-0,9592
-1
-1,3241
-1,5
-1,7499
-2
-2
-2,4510
-2,5 -2,7324
-3
-3
-3,5
-4
-4
-4,5
Series1

Operation Mangement Chapter 3

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Operation Mangement Chapter 3

Uploaded by

Copyright:

Available Formats

Operations Management: Processes and

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

• Deciding What to Forecast

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

• For any forecasting method, it is important to measure the

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Cumulative sum of Standard deviation

Mean Squared Error Mean Absolute Percent Error

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Past period 1 39 41 −2 2 4 5.128%

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Mean absolute deviation:

Mean absolute percent error:

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

• Other methods (casual, time-series, and trend projection

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

We used POM for Windows to determine the best values

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Figure 8.4 Linear

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

• Forecast for month 6:

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

• Specifically, the forecast for period t + 1 can be

Sum of last n demands Dt + Dt −1 + Dt −2 +    + Dt −n+1

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

b. If the actual number of patient arrivals in week 4 is 415,

a. The moving average forecast at Week Patient Arrivals

c. The forecast for week 5 requires the actual arrivals from

In the weighted moving average method, each historical

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

• A sophisticated weighted moving average that calculates

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

• The emphasis given to the most recent demand levels

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

a. Reconsider the patient arrival data in Example 14.3. It is

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

a. To obtain the forecast for week 4, using exponential

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

b. The forecast error for week 4 is

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

• A trend in a time series is a systematic increase or

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

• Multiplicative seasonal method

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Multiplicative seasonal method

The manager of the Stanley Steemer carpet cleaning

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Copyright © 2019 Pearson Education, Ltd. All Rights Reserved

Average Seasonal Factor Quarterly Forecasts