Upload
others
View
0
Download
0
Embed Size (px)
Citation preview
Page 1
Strategic Planning for
Logistics and Supply Chain
School of Engineering
The University of the Thai
Chamber of Commerce
Page 2
Forecasting
School of Engineering
The University of the Thai Chamber of Commerce
Page 3
Agenda
• What is forecast?
• Elements of good forecasts
• The necessary steps in preparing a forecast
• Basic forecasting techniques
• How to monitor a forecast
Page 4
Page 5
Page 6
How will demand grow ?Long time frames over which Boeing must plan.
Boeing 737
Boeing 717
Mcdonell-Douglas11
Page 7
Page 8
Boeing Long-term capacity decisions
Page 9
Page 10
Page 11
Motto in OM class
• It’s an old story, but an instructive note: Two shoe
salesmen arrive on a primitive island where no one
wears shoes. One cables his head office saying “No
business. Shoes not worn”, the other sends a
different message “Send more shoes. No
competition.”
John F. Kenedy
Page 12
1. Introduction
• Have you ever forecast??
• How much food and drink will I need for the party?
• Will I get the job?
• Which team will be a world champion in 2014?
To make these forecasts,
• One is current factors or conditions.
• The other is past experience in a similar situation.
Page 13
1. Introduction
• Forecasting are the basis for budgeting and
planning for capacity, sales, production and
inventory, personnel, purchasing, and more.
• Forecast play an important role in the planning
process.
• Forecasts affect decisions and activities throughout
an organization, in accounting, finance, human
resources, marketing, MIS, as well as operations,
and other parts of an organization.
Page 14
Accounting Cost/profit estimates
Finance Cash flow and funding
Human Resources Hiring/recruiting/training
Marketing Pricing, promotion, strategy
MIS IT/IS systems, services
Operations Schedules, MRP, workloads
Product/service design New products and services
2.1 Uses of Forecasts
Page 15
2. FORECAST:
• There are two methods for forecasting.
– Plan the system (involves long term plan about the types of products and service to offer).
– Plan to use the system (involves short and intermediate term plan such as planning inventory , workforce levels, planning purchasing, budgeting and scheduling).
Page 16
Page 17
การออกแบบศนูยก์ระจายสนิคา้ของจงัหวดัพษิณุโลกก
Page 18
Page 19
• Assumes causal systempast ==> future
• Forecasts rarely perfect because of randomness
• Forecasts more accurate forgroups vs. individuals
• Forecast accuracy decreases as time horizon increases
I see that you will
get an A this semester.
2.1 Features Common to all Forecasts
Page 20
Forecasting time horizons
- Short-range forecast (not more than one year; Planning purchasing, Job scheduling, Workforce levels and so on)
- Medium-range forecast ( 3 months to 3 years; Production planning and budgeting, Cash budgeting)
- long-range forecast (more than 3 years; planning for new products, Capital expenditures, Facility location and R&D
Page 21
The influence of product life cycle (PLC)
1 Introduction
2 Growth
3 Maturity
4 Decline
Page 22
Page 23
3. Elements of a Good Forecast
Timely
AccurateReliable
Written
Page 24
4. Steps in the Forecasting Process
Step 1 Determine purpose of forecast
Step 2 Select the items to be forecasted
Step 3 Establish a time horizon
Step 4 Select a forecasting technique
Step 5 Gather and analyze data
Step 6 Monitor the forecast
“The forecast”
Step 7 Validate and Implement the results
Page 25
5. Types of Forecasts
• Judgmental - uses subjective inputs
• Time series - uses historical data assuming the future will be like the past
• Associative models or Casual Model – use equation that consists of one or more explanatory variables to predict the future. For example, demand for paint might be related to variables such as the price per gallon and the amount spent on advertising, as well as specific characteristics of the paint.
Page 26
6. Judgmental Forecasts
• Executive opinions
• Sales force opinions
• Consumer surveys
• Outside opinion
• Delphi method
– Opinions of managers and staffs
– Achieves a consensus forecast
Page 27
7. Time Series Forecasts
• is a time-ordered sequence of observations taken at
regular intervals.
• The data may be measurements of demand, earnings,
profits, shipments, accidents, output and productivity.
• Trend - long-term movement in data
• Seasonality - short-term regular variations in data
• Cycle – wavelike variations of more than one year’s
duration
• Irregular variations - caused by unusual circumstances
• Random variations - caused by chance (Bird Flu)
Page 28
7.1 Forecast Variations
Trend
Irregular
variatio
n
Seasonal variations
90
89
88
Cycles
Page 29
7.2 Naive Forecasts
Uh, give me a minute....
We sold 250 wheels last
week.... Now, next week
we should sell....
The forecast for any period equals
the previous period’s actual value.
Page 30
• Simple to use
• Virtually no cost
• Quick and easy to prepare
• Data analysis is nonexistent
• Easily understandable
• Cannot provide high accuracy
• Can be a standard for accuracy
7.2 Naïve Forecasts
Page 31
• Stable time series data
– F(t) = A(t-1)
• Seasonal variations
– F(t) = A(t-n)
• Data with trends
– F(t) = A(t-1) + (A(t-1) – A(t-2))
7.2 Uses for Naïve Forecasts
Page 32
7.2 Naïve Methods
• Uses a single previous value of a time series as the basis
of a forecast
Period Actual Change from previous value Forecast
t-1 50
t 53 +3
t+1 53+3 = 56
Page 33
7.3 Techniques for Averaging
• Generate forecasts that reflect recent values of a
time series.
• Work best when a series tends to vary around an
average
– Moving average
– Weighted moving average
– Exponential smoothing
Page 34
7.3.1 Moving average
• Uses a number of the most recent actual data values in
generating a forecast.
n
A
MAF
n
1i
i
nt
i = an index that corresponds to periods
n = number of periods in the moving average
Ai = actual value in period i
MA = Moving Average
Ft = Forecast for period t
Page 35
Example 1
• Compute a three period
moving average forecast
given demand for
shopping carts for the last
five periods.
Period Age Demand
1 5 42
2 4 40
3 3 43
4 2 40
5 1 41
33.413
414043F6
403
3941407
FIf actual demand in period 6
turns out to be 39. What is F7 ?
Page 37
Page 38
7.3.2 Weighted Moving Average
• A weighted average is similar to a moving average,
except that it assigns more weight to the most
recent values in a time series.
• For instance, the most recent value might be
assigned a weight of .40, the next most recent value
a weight of .30, the next after that a weight of .20,
and the next after that a weight of .10.
• That weights sum to 1.00, and that the heaviest weights are assigned to the most recent values.
Page 39
7.3.2 Weighted Moving Average
a) Compute weighted average forecast using a weight .4 for the most
recent period, .3 for the next most recent, .2 for the next, and .1 for
the next.
b) If the actual demand for period 6 is 39, forecast demand for period 7
using the same weights as in part a.
Period Demand
1 42
2 40
3 43
4 40
5 41
Page 40
7.3.2 Weighted Moving Average
2.40)43(1.)40(2.)41(3.)39(4.F.b
0.41)40(1.)43(2.)40(3.)41(4.F.a
7
6
Note that if four weights are used, only the four most recent demands are used to prepare the forecast.
Page 41
• The weighted average is more reflective of the most
recent occurrences.
• The choice of weights is somewhat arbitrary and
generally involves the use of trial and error to find a
suitable weighting scheme.
7.3.2 Weighted Moving Average
Page 42
Page 43
7.3.3 exponential smoothing
• Exponential smoothing is a sophisticated weighted
averaging method that is still relatively easy to use
and understand. Each new forecast is based on the
previous forecast plus a percentage of the
difference between that forecast and the actual
value of the series at that point.
Page 44
7.3.3 Exponential Smoothing
Forecast) Previous - (Actual forecast Previous forecast Next
α represents a percentage of the forecast error.
Therefore, each new forecast is equal to the previous forecast plus a
percentage of the previous error.
Suppose the previous forecast was 42 units, actual demand was 40
units, and α = .10. the new forecasts
F = 42 + .10(40-42) = 41.8
Then if the actual demand turns out to be 43, the next forecast would
be??
)FA(FF 1t1t1tt
Ans. 41.92
Page 45
7.3.3 Exponential Smoothing
• An alternate form of formula reveals the weighting of the
previous forecast and the latest actual demand:
• For example:
1t1tt
1t1t1tt
AF)1(F
)FA(FF
1t1tt
1t1t1tt
A1.0F)9.0(F
)FA(10.0FF
F = 42 + .10(40-42) = (0.9)(42) + (.10)(40) = 41.8
Page 46
Example 2
• The following table illustrates two series of forecasts
for a data set and the resulting error for each
period. One forecast uses α = .10 and one uses α =
.40. The following figure plots the actual data and
both sets of forecasts.
Page 47
Period Actual Alpha = 0.1 Error Alpha = 0.4 Error
1 42
2 40 42 -2.00 42 -2
3 43 41.8 1.20 41.2 1.8
4 40 41.92 -1.92 41.92 -1.92
5 41 41.73 -0.73 41.15 -0.15
6 39 41.66 -2.66 41.09 -2.09
7 46 41.39 4.61 40.25 5.75
8 44 41.85 2.15 42.55 1.45
9 45 42.07 2.93 43.13 1.87
10 38 42.36 -4.36 43.88 -5.88
11 40 41.92 -1.92 41.53 -1.53
12 41.73 40.92
Example 2 - Exponential Smoothing
Page 48
Picking a Smoothing Constant
35
40
45
50
1 2 3 4 5 6 7 8 9 10 11 12
Period
De
ma
nd .1
.4
Actual
Page 49
7.3.3 Exponential Smoothing
• The closer α is to zero, the slower the forecast will be to
adjust to forecast errors. (the greater the smoothing,
emphasis the previous data)
• The closer the value of α is to 1.00, the greater the
responsiveness and the less the smoothing. (emphasis the present data )
nt
n
tttt AAAAF )1(...)1()1( 3
2
21
Page 50
7.4 Techniques for trend
• Develop an equation that will suitably describe trend
• The trend component may be linear, or it may not.
• Two important techniques that can be used to
develop forecasts
– Trend equation
– Extension of exponential smoothing
Page 51
7.4 Common Nonlinear Trends
Parabolic
Exponential
Growth
Figure 3.5
Page 52
7.5 Linear Trend Equation
• Ft = Forecast for period t
• t = Specified number of time periods
• a = Value of Ft at t = 0
• b = Slope of the line
Ft = a + bt
0 1 2 3 4 5 t
Ft
Page 53
7.5 Trend equation
The coefficients of the line, a and b, can be computed from
historical data using these components.
tb-yor n
tbya
)t(tn
yttynb
22
n = number of periods
y = value of the time series
Page 54
Linear Trend Equation Example
t y
Week t2
Sales ty
1 1 150 150
2 4 157 314
3 9 162 486
4 16 166 664
5 25 177 885
t = 15 t2 = 55 y = 812 ty = 2499
(t)2 = 225
Page 55
Linear Trend Calculation
y = 143.5 + 6.3t
a =812 - 6.3(15)
5=
b =5 (2499) - 15(812)
5(55) - 225=
12495-12180
275 -225= 6.3
143.5
Page 56
Cell phone sales for a
California-based firm over the
last 10 weeks are shown in the
following table. Plot the data,
and visually check to see if a
linear trend line would be
appropriate. Then determine
the equation of the trend line,
and predict sales for weeks 11
and 12.
Example 3
Week Unit Sales
1 700
2 724
3 720
4 728
5 740
6 742
7 758
8 750
9 770
10 775
Page 57
a. A plot suggests that a linear trend line would be appropriate:
Example 3
unit sales
660
680
700
720
740
760
780
800
1 2 3 4 5 6 7 8 9 10 11 12
week
sa
les
Page 58
Week (t) Unit Sales
(y)
Ty
1 700 700
2 724 1448
3 720 2160
4 728 2912
5 740 3700
6 742 4452
7 758 5306
8 750 6000
9 770 6930
10 775 7750
55 7407 41358
b.
699.40 10
)55(51.7407,7a
51.7825
195,6
)55(55)385(10
)407,7)(55()358,41(10b
Example 3
Thus the trend line is
t51.740.699yt
tb-yor n
tbya
)t(tn
yttynb
22
Page 59
c. Substituting values of t into this equation, the forecasts
for the next two periods are:
Example 3
52.789)12(51.740.699y
01.782)11(51.740.699y
12
11
Page 60
unit sales
660
680
700
720
740
760
780
800
1 2 3 4 5 6 7 8 9 10 11 12
week
sa
les
Example 3
d. For purposes of illustration, the original data, the trend line,
and the two projections (forecasts) are shown on the following
graph.
Page 61
7.6 Associative Forecasting
• Associative techniques rely on identification of related variables that can be used to predict values of the variable of interest.
• Predictor variables - used to predict values of variable interest
• Regression - technique for fitting a line to a set of points
• Least squares line - minimizes sum of squared deviations around the line
Page 62
7.7 Linear Model Seems Reasonable
A straight line is fitted to a set of sample points.
0
10
20
30
40
50
0 5 10 15 20 25
X Y
7 15
2 10
6 13
4 15
14 25
15 27
16 24
12 20
14 27
20 44
15 34
7 17
Computed
relationship
Page 63
8. Forecast Accuracy
• Error = actual value - predicted value
• Mean Absolute Deviation (MAD)
– Average absolute error
• Mean Squared Error (MSE)
– Average of squared error
• Mean Absolute Percent Error (MAPE)
– Average absolute percent error
Page 64
8.1 MAD, MSE, and MAPE
MAD =Actual forecast
n
MSE =Actual forecast)
-1
2
n
(
MAPE =Actual forecas
t
n
/ Actual*100)(
Page 65
Example 4
Period Actual Forecast (A-F) |A-F| (A-F)^2 (|A-F|/Actual)*100
1 217 215 2 2 4 0.92
2 213 216 -3 3 9 1.41
3 216 215 1 1 1 0.46
4 210 214 -4 4 16 1.90
5 213 211 2 2 4 0.94
6 219 214 5 5 25 2.28
7 216 217 -1 1 1 0.46
8 212 216 -4 4 16 1.89
-2 22 76 10.26
MAD= 2.75
MSE= 10.86
MAPE= 1.28
Page 66
Example 4: Solution
• MAD = 22/8 = 2.75
• MSE = 76/(8-1) = 10.86
• MAPE = 10.26%/8 = 1.28%
Page 67
9. Controlling the Forecast
• Control chart
– A visual tool for monitoring forecast errors
– Used to detect non-randomness in errors
• Forecasting errors are in control if
– All errors are within the control limits
– No patterns, such as trends or cycles, are present
Page 68
9.1 Control Chart
• S =
• UCL :
• LCL :
MSE
MSEz
MSEz
• MSE = 2
• S =
• UCL :
• LCL :
41.1MSE
82.2MSEz
82.2MSEz
+2.82
-2.82
Page 69
10. Sources of Forecast errors
• Model may be inadequate
• Irregular variations
• Incorrect use of forecasting technique
Page 70
11. Tracking Signal or Control Chart
Tracking signal =(Actual-forecast)
MAD
•Tracking signal
–Ratio of cumulative error to MAD
Bias – Persistent tendency for forecasts to be
Greater or less than actual values.
Page 71
12. Choosing a Forecasting Technique
• No single technique works in every situation
• Two most important factors
– Cost
– Accuracy
• Other factors include the availability of:
– Historical data
– Computers
– Time needed to gather and analyze the data
– Forecast horizon
Page 72
13. Forecast factors, by range of forecast
Factor Short Range Intermediate
Range
Long Range
1. Frequency Often Occasional Infrequent
2. Level of Aggregation Item Product family Total output, type of
product/service
3. Type of model Smoothing,
projection,
regression
Smoothing,
projection,
regression
Managerial judgment
4. Degree of management
involvement
Low Moderate High
5. Cost per forecast Low Moderate high
Page 73
Problem 1
1. The appropriate naïve approach
2. A three period moving average and five period
3. A weighted average using weights of .50 (most recent), .30, and .20
4. Exponential smoothing with a smoothing constant of .40
Period Number of Complaints
1 60
2 65
3 55
4 58
5 64
Page 74
Solution:
1. The values are stable. Therefore, the most recent
value of the series becomes the next forecast: 64
2. MA3 = (55+58+64)/3 = 59
MA5 = (60+65+55+58+64)/5 = 60.4
3. F = .20(55)+.30(58)+.50(64) = 60.4
Page 75
Solution:
Period Number of
complaints
Forecast calculations
1 60
2 65 60
3 55 62 60+.40(65-60) = 62
4 58 59.2 62+.40(55-62) = 59.2
5 64 58.72 59.2 + .40(58-59.2) = 58.72
6 60.83 58.72+.40(64-58.72) = 60.83
Page 76
Problem 2:
• Plot the data on a graph,
and verify visually that a
linear trend line is
appropriate. Develop a line
trend equation for the
following data. Then use
the equation to predict the
next two value of the
series
Period Demand
1 44
2 52
3 50
4 54
5 55
6 55
7 60
8 56
9 62
Page 77
Solution 2:
0
10
20
30
40
50
60
70
0 2 4 6 8 10 12
Period
Dem
an
d
Page 78
Solution 2:
Period (t) Demand (y) Ty
1 44 44
2 52 104
3 50 150
4 54 216
5 55 275
6 55 330
7 60 420
8 56 448
9 62 558
448 2545
Page 79
Solution 2:
45.47 9
)45(75.1488a
75.1)45(45)285(9
)488)(45()545,2(9b
Thus the trend line is
72.64)11(75.147.45F
97.62)10(75.147.45F
t75.147.45F
11
10
t
tb-yor n
tbya
)t(tn
yttynb
22
Page 80
Problem 3
• The manager of a large manufacturer of industrial pumps must chose
between two alter active forecasting techniques. Both techniques have
been used to prepare forecasts for a six-month period. Using MAD as
a criterion, which technique has the better performance record?
Forecast
Month Demand Technique 1 Technique 2
1 492 488 495
2 470 484 482
3 485 480 478
4 493 490 488
5 498 497 492
6 492 493 493
Page 81
Solution 3
Page 82
Solution 3:
• Technique 1 is superior in this comparison because
its MAD is smaller, although six observations would
generally be too few on which to base a realistic
comparison.
Page 83
Problem 4:
• Given the demand data that follow, prepare a naïve forecast
for periods 2 through 10. Then determine each forecast
error, and use those values to obtain 2s control limits. If
demand in the next two periods turns out to be 125 and 130, can you conclude that the forecasts are in control?
Period 1 2 3 4 5 6 7 8 9 10
Demand 118 117 120 119 126 122 117 123 121 124
Page 84
Solution 4:
Period Demand Forecast Error Error square
1 118 - - -
2 117 118 -1 1
3 120 117 3 9
4 119 120 -1 1
5 126 119 7 49
6 122 126 -4 16
7 117 122 -5 25
8 123 117 6 36
9 121 123 -2 4
10 124 121 3 9
6 150
Page 85
Solution 4:
33.419
150
1n
errors
2
n = Number of errors
The control limits are 2(4.33) = +/-8.66
The forecast for period 11 was 124. demand turned out to be 125, for an
error of 125-124 = 1. this is within the limits of +/-8.66. If the next demand
is 130 and the naïve forecast is 125, the error is +5. again, this is within
the limits, so you cannot conclude the forecast is not working properly.
With more values at least five or six you could plot the errors to see
whether you could detect any patterns suggesting the presence of non-randomness.
Page 86
Problem 5:
5. National mixer Inc. sell can
openers. Monthly sales for
a seven-month period
were as follows:
Month Sales
(000 units)
Feb 19
Mar 18
Apr 15
May 20
Jun 18
Jul 22
Aug 20
Page 87
Problem 5
a. Plot the monthly data
b. Forecast September sales volume using each of the
following:
a. A linear trend equation.
b. A five-month moving average
c. Exponential smoothing with alpha = 0.20, assuming a March
forecast of 19(000).
d. The naïve approach
e. A weighted average using .60 for August, .30 for July, and .10
for June.
c. Which method seems least appropriate? Why?
Page 88
Problem 6
6. Freight car loadings over a 12 year period at a busy port are
Week Ton Shipped Week Ton Shipped Week Ton Shipped
1 405 8 433 15 466
2 410 9 438 16 474
3 420 10 440 17 476
4 415 11 446 18 482
5 412 12 451
6 420 13 455
7 424 14 464
Page 89
Problem 6:
a. Determine a linear trend line for freight car
loadings.
b. Use the trend equation to predict loadings for
weeks 20 and 21.
c. The manager intends to install new equipment
when the volume exceeds 800 loadings per week.
assuming the current trend continues, the loading
volume will reach that level in approximately what
week?
Page 90
Problem 7:
7. Two different forecasting techniques were used to forecast
demand fore cases of bottled water. Actual demand and the two sets of forecasts are as follows:
Forecast
Period Demand Technique 1 Technique 2
1 68 66 66
2 75 68 68
3 70 72 70
4 74 71 72
5 69 72 74
6 72 70 76
7 80 71 78
8 78 74 80
Page 91
Problem 7:
a) Compute MAD for set of forecasts. Given your results,
which forecast appears to be more accurate? Explain
b) Compute the MSE for each set of forecasts. Given your
results, which forecast appears to be more accurate?
c) In practice, either MAD or MSE would be employed to
compute forecast errors. What factors might lead a
manager to choose one rather than the other?
d) Compute MAPE for each data set. Which forecast appears to be more accurate?
Page 92