Revision

1. Performance measures of different Production systems

• Production rate
• Lead time
• WIP
• Utilization

2. Difference between takt time and cycle time?

3. Cellular manufacturing and flexible manufacturing system?

1
Forecasting

2
 Schematic model of Production system
(Demand)
(4. Facility Location and 5. layout)

Forecasting 7. SCM

Transformation
Activities
Personnel Legal and
E social
1. Product design and 3.
N E
process planning
V O Marketing N
I I 9. Production control U and public V
N Material and Goods and T I
R Engineering relations
P equipment Services P R
O Aggregate Material Scheduling
N U Requirement U O
planning and control
M T planning T N
E S S M
N Finance E
10. Maintenance Accounting N
T
T

8. Inventory 2. Quality Cost

Control Control Control

Feed back
3
4
Walt Disney Parks and
Resorts, a visible global
leader.

5
 Agenda
• Global Company Profile: Walt Disney Parks & Resorts
• What is Forecasting?
• The Strategic Importance of Forecasting
• Seven Steps in the Forecasting System
• Forecasting Approaches
• Time-Series Forecasting
• Associative Forecasting Methods: Regression and Correlation Analysis
• Monitoring and Controlling Forecasts
• Forecasting in the Service Sector

6
 Forecasting Provides a Competitive
Advantage for Disney
► Global portfolio includes parks in Hong Kong, Paris, Tokyo,
Orlando(Florida), and Anaheim(California).
► Revenues are derived from people – how many visitors and how they spend
their money
► Daily management report contains only the forecast and actual attendance
at each park.
► Disney generates daily, weekly, monthly, annual, and 5-year
forecasts
► Forecast used by labor management, maintenance, operations,
finance, and park scheduling
► Forecast used to adjust opening times, rides, shows, staffing levels,
and guests admitted

7
► 20% of customers come from outside the USA, Economic model includes
gross domestic product, cross-exchange rates, arrivals into the USA etc.
► A staff of 35 analysts and 70 field people survey 1 million park guests,
employees, and travel professionals each year
► Inputs to the forecasting model include airline specials, Federal Reserve
policies, Wall Street trends, vacation/holiday schedules for 3,000 school
districts around the world
► Average forecast error for the 5-year forecast is 5%
► Average forecast error for annual forecasts is between 0% and 3%

8
 What is Forecasting?
• Forecasting is the prediction, projection or estimation of the occurrences of
uncertain future events or level of activity.

• Events may be sales, breakdowns, demand rejections etc.

• It is used for predicting Demand, Revenues, Costs, Profits, Prices,

Technological changes, Environment problems, Rainfall, etc.

• Underlying basis of all business decisions

 Production

 Inventory
 Personnel
 Facilities

9
Forecast is one input to many types of planning and
control

Policy decisions (economic, social,

political, technological conditions)

Product design (product lines, services and

market)

Forecasting Process decision (process and methods)

Plant decision (facility location and layout)

Operations decisions (output

scheduling and control)

Master forecasting 10
 Functional Forecasting

Financial planning (financial aggregate, cash

flow, balance sheets, income statement)

Market planning (product lines, pricing,

Forecasting
and promotion

Production planning (aggregate output levels)

Master scheduling (product output levels)

11
Forecasting usually involves the following
considerations:

1. Item to be forecasted (products, product groups, assemblies, etc.),

2. Top down or bottom up forecasting ,
3. Forecasting techniques (quantitative or qualitative model),
4. Units of measure (Rs, units, weights, etc.),
5. Time interval (weeks, months, quarters, etc.),
6. Forecast horizons (how many time intervals to include),
7. Forecasting components (levels, trends, seasonal, cycles and
random variations) ,
8. Forecast accuracy (error measurement),
9. Exception reporting and special situations
10. Revision of forecasting model parameters

12
 Forecasting Time Horizons
1. Short-range forecast
► Up to 1 year, generally less than 3 months
► Planning Purchasing, job scheduling, workforce levels, job
assignments, production levels etc.

2. Medium-range forecast
► 3 months to 3 years
► Sales and production planning, budgeting etc.

3. Long-range forecast
► 3+ years
► New product planning, facility location, research and development
capital expenditure etc.
13
 Distinguishing Differences
1. Medium/long range forecasts deal with more comprehensive issues
and support management decisions regarding planning and
products, plants and processes.

2. Short-term forecasting usually employs different methodologies

than longer-term forecasting

3. Short-term forecasts tend to be more accurate than longer-term

forecasts

14
 Influence of Product Life Cycle

Introduction – Growth – Maturity – Decline

► Introduction and growth require longer forecasts than maturity and decline

► As product passes through life cycle, forecasts are useful in projecting

► Staffing levels
► Inventory levels
► Factory capacity

15
 Product Life Cycle
Introduction Growth Maturity Decline

Best period to Practical to change Poor time to Cost control

increase market price or quality change image, critical
share image price, or quality
Company Strategy/Issues

R&D engineering is Strengthen niche Competitive costs

critical become critical
Defend market
position Drive-through
Internet search engines restaurants
DVDs
Xbox 360
iPods
Boeing 787

Sales
3D printers

3-D game Analog

Electric vehicles TVs
players

16
 Product Life Cycle
Introduction Growth Maturity Decline
Product design and Forecasting critical Standardization Little product
development Product and Fewer product differentiation
critical process reliability changes, more Cost
Frequent product Competitive minor changes minimization
and process
OM Strategy/Issues

product Optimum capacity Overcapacity in

design changes improvements and the industry
Increasing stability
Short production options of process Prune line to
runs Increase capacity eliminate items
Long production
High production Shift toward runs not returning
costs product focus good margin
Product
Limited models Enhance improvement and Reduce
Attention to quality distribution cost cutting capacity

17
 Types of Forecasts
1. Economic forecasts
► Address business cycle – inflation rate, money supply, housing
starts and other planning indicators.
2. Technological forecasts
► Predict rate of technological progress.
► Impacts development of new products.
3. Demand/ Sales forecasts
► Predict sales of existing products and services.
► Demand driven forecasts focus on rapidly identifying and tracking
customer desires.
► It drive a company’s production, capacity and scheduling systems
and serves as input to financial, marketing and personal planning.

18
 Strategic Importance of Forecasting
► Supply-Chain Management – Good supplier relations, advantages in
product innovation, cost and speed to market

► Human Resources – Hiring, training, laying off workers

► Capacity – Capacity shortages can result in undependable delivery, loss of

customers, loss of market share

19
 Seven Steps in Forecasting
1. Determine the use of the forecast
2. Select the items to be forecasted
3. Determine the time horizon of the forecast
4. Select the forecasting model(s)
5. Gather the data needed to make the forecast
6. Make the forecast
7. Validate and implement results

20
 The Realities!
► Forecasts are seldom perfect, unpredictable outside factors may impact the
forecast

► Most techniques assume an underlying stability in the system

► Product family and aggregated forecasts are more accurate than individual
product forecasts

21
 Sales Forecasting
• Sales forecasts are used to establish product levels, facilitate
scheduling, set inventory levels, determine manpower loading, make
purchasing decisions, establish sales conditions – pricing and
advertising, and financial planning – cash budgeting and capital
budgeting.
• Generally, sales forecast is used to estimate the demand of
independent items.
• Many environmental factors influence the demand for products and
services of an organisation.
• Some major environmental factors are
– General business conditions and state of the economy.
– Competitor actions and reactions
– Governmental legislative actions
– Marketplace trend
• Product life cycle
• Style and fashion
• Changing consumer demands
– Technological innovations
22
• Presence of randomness preclude a perfect forecast

• Forecast for groups of items tend to be more accurate than forecast for individual items

• Error potential increases as time horizon of a forecast increases

• We are interested in estimating the level of future demand. Statistical techniques are used
to forecast.

• Statistical methods use historical (past) data

• All statistical forecasting techniques assume to some extent that forces that have existed
in the past will persist in the future.

• New product demand (with little or no history of past demand) rely more on subjective
phenomenon and solicitation of opinions.

• Direct survey approach – asking prospective customers of their buying interest

• Indirect survey approach – information from salesmen, wholesalers, area managers, etc

• Comparison with substitute or comparable products Limited market test of the new
product

23
 Forecasting Approaches
Qualitative Methods
► Used when situation is vague and little data exist
► New products
► New technology
► Involves intuition, experience
► e.g., forecasting sales on Internet

24
Quantitative Methods
► Used when situation is ‘stable’ and historical data exist

► Existing products

► Current technology

► Involves mathematical techniques

► e.g., forecasting sales of color televisions

25
Basic demand forecasting models
• Time series analysis, soliciting opinions, economic indicators
and econometric models

• These are short range forecasting models

• Generally, these forecasts give starting point for making the

final forecast

• Final forecast usually requires an additional input in the form

of judgment, intuition, and experience and requires periodic
review

Note on economic indicators and econometric models

Economic Indicators
• Knowledge of one variable is used to predict the value of
another (prediction by association)

26
• Certain economic indicators are
– Gross domestic product (GDP), Personal income, Bank deposits, Freight car loadings, etc

• One or more of these indicators have relationship with the forecast

variable
Econometric Models:

• Involves a set of simultaneous equations that explains the interactions of

variables involved in a business situation

• Attempt to show the relationships between relevant variables such as

supply, demand, prices and purchasing power of the consume

27
 Overview of Qualitative Methods
1. Jury of executive opinion
► Pool opinions of high-level experts, sometimes augment by
statistical models
2. Delphi method
► Panel of experts, queried iteratively
3. Sales force composite
► Estimates from individual salespersons are reviewed for
reasonableness, then aggregated
4. Market Survey
► Ask the customer

28
 Jury of Executive Opinion
► Involves small group of high-level experts and managers

► Group estimates demand by working together

► Combines managerial experience with statistical models

► Relatively quick

29
 Delphi Method
► Iterative group
Decision Makers
process, continues (Evaluate responses
and make decisions)
until consensus is
Staff
reached (Administering
survey)
► 3 types of participants

► Decision makers
► Staff
► Respondents
Respondents
(People who can make
valuable judgments)
30
 Sales Force Composite
► Each salesperson projects his or her sales
► Combined at district and national levels
► Sales reps know customers’ wants
► May be overly optimistic

31
 Market Survey
► Ask customers about purchasing plans

► Useful for demand and product design and planning

► What consumers say, and what they actually do may be different

► May be overly optimistic

32
 Overview of Quantitative
Approaches

1. Naive approach
2. Moving averages
3. Exponential Time-series
smoothing models
4. Trend projection
5. Linear regression Associative
model

33
 Time-Series Analysis/Forecasting
► Time series analysis predict future demand from past interval data.
► A time series is a set of time-ordered observations on a variable during
successive and equal time periods.
Period Jan Feb Mar Apr May Jun
Demand( in units) 75 70 82 76 87 80

► The above table shows a time series that shows the past demand in
successive and equal interval of time.
Period Jan Mar Apr May July Sep
Demand( in units) 75 70 82 76 87 80

► The above table is not representing a time series as it is not showing

demand in equal interval of time.
34
 Time-Series Components

Trend Cyclical

Seasonal Random

35
 Raw data

Trend

Seasonal

Cyclic

Random

Time (years)

Various components of a time series

36
 Components of Demand
Trend
component
Demand for product or service

Seasonal peaks

Actual demand
line

Average demand
over 4 years

Random variation
| | | |
1 2 3 4
Time (years)

37
 Trend Component
► Trend identify the rate of growth or decline of a series over time (Long-
term historical pattern of demand over time)

► Changes due to population, technology, age, culture, etc.

38
 Seasonal Component
► Seasonal variations consists of annually recurring movements above and
below the trend line. It is due to weather, custom etc.
– Demand fluctuates in a repetitive pattern from year to year
– Seasonal periodic peaks and valleys should occur at the same time
every year
– Seasonal variations should be of larger magnitude than the random
variations

► E.g. restaurant and salons for weekly seasons, a/c and refrigerator yearly.

PERIOD LENGTH “SEASON” LENGTH NUMBER OF “SEASONS” IN PATTERN

Week Day 7
Month Week 4 – 4.5
Month Day 28 – 31
Year Quarter 4
Year Month 12
Year Week 52

39
 Cyclical Component
• Cyclical variations are long term oscillations or swings about a trend line .
It is the pattern in the data that occurs every several years. Repeating up
and down movements.

• The cycles may or may not be periodic, but they often are the result of
business cycles of expansion and contraction of economic activity over a
number of years. It is affected by business cycle, political, and economic
factors

• Business cycles may be due to one or more of the following: prosperity,

recession, depression and recovery

• The cycles may vary with respect to the time of occurrence, the length of
the phases, and the amplitude of the fluctuations. It has multiple years of
duration and often has a causal or
associative relationships.

40
 Random Component
• Random variations have no particular pattern and usually are without
specific assignable cause. They are erratic, unsystematic, ‘residual’
fluctuations.

• They represent all influences not included in trend, seasonal, and cyclical
variations. They occur due to random variation or unforeseen events.

• Erratic occurrence may be isolated and removed from the data, but there
are no general techniques for doing so. Averaging process will help to
eliminate its influence.

• Random variations are often referred to as noise, residuals, or irregular

variations. They are Short duration and nonrepeating .

M T W T F
41
The time series contains interactive components. The models
representing interactive components of demand are classified as
– Multiplicative model
– Additive model
– Mixed model (partially additive, partially multiplicative)

The demand in period (t) for a multiplicative model is represented as

Demand = (Trend) (seasonal) (cycle) (random)

Dt = b × F × c × et

The demand in period (t) for a additive model is represented as

Demand = level + trend + seasonal + cyclic + random

Dt = a + b × t + Ft + Ct + et

42
Generally, demand process can be modelled as

Dt = a + et – (level model – additive type)

Dt = a + b × t + et – (trend model – additive type)

Dt = (a + b × t) × Ft + et – (mixed model type- trend part is additive and

seasonal part is in multiplicative in form)

Some Notations:
• Dt – Actual demand for the period t

• ft - forecast for the period t

43
 Naive Approach/ Last period demand
► Assumes demand in next period is the same as demand in most recent
period
► e.g., If January sales were 68, then February sales will be 68

► Sometimes cost effective and efficient

► Can be good starting point

44
 Simple Moving Average Method
• This method is used to represent a demand process of type
Dt = a + et
That is, the demand is represented as a level with random noise.

• Parameter a is not really known and is subjected to random changes from

time to time.

• Using the simple moving average procedure we can get an estimate for a
and it can be get updated as time progresses.

• MA is useful if we assume that market demand will stay fairly steady over
a time and it is a series of arithmetic means. It is used if little or no trend is
present

Moving average =
åDemand in previous N periods
N
► N is the number of periods in moving average.
45
 Estimating procedure (updating procedure)
• The procedure involves the determination of average of demand of last N
periods.
• As new period demand observation is available, the old period demand data
is removed from average calculation.
• Number of periods considered for average calculation is same but, demand
data considered for the calculation is different at different time periods.
• This way of estimation is actually an updating procedure also.

MAt,N = (Dt + Dt-1 + Dt-2 + …. Dt-N+1)/N

Where, N is the period of moving average, Dt is the actual demand at period t
and MAt,N is the moving average at period t based on demand of N periods.
• The estimate of a, as of the end of the period t is represented as

ât = MAt,N
• This estimate of a, results from minimizing the sum of squares of error over
the preceding N period.

46
• A slightly simple updating procedure for this method is

• ât minimizes the standard error,

Forecast equations are

• ft+1 = MAt,N
• ft+n = MAt,N (forecast for n period ahead)

47
1. Harish garden supply wants a 3-month moving average forecast, including a
forecast for next january, for shed sales.

Data
MONTH ACTUAL SHED SALES
January 10
February 12
March 13
April 16
May 19
June 23
July 26
August 30
September 28
October 18
November 16
December 14

48
 Moving Average Example

MONTH ACTUAL SHED SALES 3-MONTH MOVING AVERAGE

January 10
February 12
March 13
April 16 (10 + 12 + 13)/3 = 11 2/3
May 19 (12 + 13 + 16)/3 = 13 2/3
June 23 (13 + 16 + 19)/3 = 16
July 26 (16 + 19 + 23)/3 = 19 1/3
August 30 (19 + 23 + 26)/3 = 22 2/3
September 28 (23 + 26 + 30)/3 = 26 1/3
October 18 (29 + 30 + 28)/3 = 28
November 16 (30 + 28 + 18)/3 = 25 1/3
December 14 (28 + 18 + 16)/3 = 20 2/3

49
 Weighted Moving Average
► Used when some trend might be present
► Older data usually less important

► Weights based on experience and intuition

(( )(
Weighted å Weight for period n Demand in period n
moving =
))
average å Weights

50
 Weighted Moving Average

MONTH ACTUAL SHED SALES 3-MONTH WEIGHTED MOVING AVERAGE

January 10
February 12
March 13
April 16 [(3 x 13) + (2 x 12) + (10)]/6 = 12 1/6
May 19
June 23
July 26
August 30
September 28
October 18
November 16
December 14

51
 Weighted Moving Average

MONTH ACTUAL SHED SALES 3-MONTH WEIGHTED MOVING AVERAGE

January 10
February 12
March 13
April 16 [(3 x 13) + (2 x 12) + (10)]/6 = 12 1/6
May 19
June 23
WEIGHTS APPLIED PERIOD
July 26 3 Last month
August 30 2 Two months ago
September 28 1 Three months ago
October 18 6 Sum of the weights
November 16this month =
Forecast for
December 3 x14
Sales last mo. + 2 x Sales 2 mos. ago + 1 x Sales 3 mos. ago
Sum of the weights

52
 Weighted Moving Average

MONTH ACTUAL SHED SALES 3-MONTH WEIGHTED MOVING AVERAGE

January 10
February 12
March 13
April 16 [(3 x 13) + (2 x 12) + (10)]/6 = 12 1/6
May 19 [(3 x 16) + (2 x 13) + (12)]/6 = 14 1/3
June 23 [(3 x 19) + (2 x 16) + (13)]/6 = 17
July 26 [(3 x 23) + (2 x 19) + (16)]/6 = 20 1/2
August 30 [(3 x 26) + (2 x 23) + (19)]/6 = 23 5/6
September 28 [(3 x 30) + (2 x 26) + (23)]/6 = 27 1/2
October 18 [(3 x 28) + (2 x 30) + (26)]/6 = 28 1/3
November 16 [(3 x 18) + (2 x 28) + (30)]/6 = 23 1/3
December 14 [(3 x 16) + (2 x 18) + (28)]/6 = 18 2/3

53
 Potential Problems With
Moving Average
► Both simple and weighted averages are effective in smoothing out sudden
fluctuations in demand pattern to provide stable estimates.

► Increasing N smooths the forecast but makes the method less sensitive to
changes in the data.

► Does not forecast trends well

► Requires extensive historical data

54
 Graph of Moving Averages
Weighted moving average
30 –

25 –
Sales demand

20 –

15 – Actual sales

10 – Moving average

5–
| | | | | | | | | | | |

J F M A M J J A S O N D
Month

55
 Simple Exponential Smoothing/ Simple
EWMA
► It is a form of weighted moving average
► In this method Weights decline exponentially
► Most recent data weighted most

► It requires smoothing constant (α)

► Which ranges from 0 to 1
► It is subjectively chosen

► It involves little record keeping of past data

56
• Underlying demand model is
Dt = at + et
where, et is normally distributed with mean zero

• A best estimates for at is the exponentially weighted smoothing average.

• The equation for simple exponential smoothing uses only two pieces of information: (1)
actual demand for the most recent period and (2) the most recent average

• Let Xt = exponentially weighted moving average for the period t

X t = α Dt + (1 - α ) X t -1

• Forecast equation is
ft+1 = Xt

• The above equation for Xt can be written as X t = X t -1 + α(Dt - X t -1 )

That is, X t = ft + α (Dt - f t )

• This equation indicates that using exponential average in one period as a forecast for the
next period; it is possible to revise the average upward or downward, depending on the
forecast error.

57
• Weights for the past data and for the initial average can be easily identified
from the equation given below.

• The weight for demand in a period k from now (t) is α (1 - α) k

• Expansion of exponentially weighted moving average equation
Xt = α Dt + (1 - α)[α Dt -1 + (1 - α) X t -2 ]
= α Dt + α (1 - α )Dt -1 + (1 - α ) 2 X t -2
= α Dt + α (1 - α)Dt -1 + (1 - α) 2 [Dt -2 + (1 - α ) X t -3 ]
= α Dt + α(1 - α )Dt -1 + α (1 - α ) 2 Dt -2 + (1 - α)3 X t -3
• If exponential average is determined for third period, the weight for the
demand of various periods and the initial average can be clearly seen from
the equation below
X 3 = α D3 + α(1 - α )D2 + α (1 - α ) 2 D1 + (1 - α )3 X 0

58
• For α = 0.2, the weight assigned for the current period demand and the
demand of past 9 periods in EWMA is as given in the table below.

• Values given in the above table are shown graphically next. This graph
shows that weight for the past demand in the current period demand
forecast is exponentially decreasing.

59
• For the demand process, the best estimate of at which minimize the
following the sum of discounted squares of residuals

where, d = a distant factor (0 < d < 1)

• The resulting estimate of at satisfies the following updating formation

aˆt = αDt + (1 - α )aˆt -1
• Average age of data in a simple EWMA is 1/α period. In a N month
moving average the average age of data is (N+1)/2
1/α = (N+1)/2
• Relationship between N (period of moving average) and α is
α = 2/(N+1)

60
Initialization
• When significant historical data exists, simply use the average demand in
the first several periods as the initial estimate of Xt.
Forecast for future periods
• ft+1 = Xt
• ft+n = Xt (forecast for n period ahead)

61
 Simple Exponential Smoothing/ Simple
EWMA
New forecast = Last period’s forecast
+  (Last period’s actual demand
– Last period’s forecast)

Ft = Ft – 1 + α(Dt – 1 - Ft – 1)

where Ft = new forecast

Ft – 1 = previous period’s forecast
α = smoothing (or weighting) constant (0 ≤ α ≤ 1)
Dt – 1 = previous period’s actual demand

62
 Exponential Smoothing Example

In January, a car dealer predicted February demand for 142 Ford Mustangs.
Actual Demand was 153 Mustangs. Using a smoothening constant choosen by
management of alpha = 0.20, the dealer wants to forecast march demand using
simple exponential smoothening model.

63
 Exponential Smoothing Example
Predicted demand = 142 Ford Mustangs

Actual demand = 153

Smoothing constant α = .20

64
 Exponential Smoothing Example

Predicted demand = 142 Ford Mustangs

Actual demand = 153
Smoothing constant  = .20

New forecast = 142 + .2(153 – 142)

65
 Exponential Smoothing Example

Predicted demand = 142 Ford Mustangs

Actual demand = 153
Smoothing constant  = .20

New forecast = 142 + .2(153 – 142)

= 142 + 2.2
= 144.2 ≈ 144 cars

66
 Effect of Smoothing Constants

▶ Smoothing constant generally .05 ≤ α ≤ .50

▶ As α increases, older values become less significant

WEIGHT ASSIGNED TO
MOST 2ND MOST 3RD MOST 4th MOST 5th MOST
RECENT RECENT RECENT RECENT RECENT
SMOOTHING PERIOD PERIOD PERIOD PERIOD PERIOD
CONSTANT ( ) (1 – ) (1 – )2 (1 – )3 (1 – )4
 = .1 .1 .09 .081 .073 .066

 = .5 .5 .25 .125 .063 .031

67
During the past eight quarters, the port of Baltimore has unloaded large
quantities of grain from ships. The ports operations manager want the teat the
use of exponential smoothening to see how well the technique works in
predicting tonnage unloaded. He guises that the forecast of grain unloaded in
the first quarter was 175 tons. Two values of alpha that is 0.10 and 0.50 is
given.

ACTUAL
TONNAGE
QUARTER UNLOADED
1 180

2 168

3 159

4 175

5 190

6 205

7 180

8 182

9 ? 68
 ACTUAL
TONNAGE FORECAST WITH
QUARTER UNLOADED FORECAST WITH  = .10  = .50
1 180 175 175

2 168 175.50 = 175.00 + .10(180 – 175) 177.50

3 159 174.75 = 175.50 + .10(168 – 175.50) 172.75

4 175 173.18 = 174.75 + .10(159 – 174.75) 165.88

5 190 173.36 = 173.18 + .10(175 – 173.18) 170.44

6 205 175.02 = 173.36 + .10(190 – 173.36) 180.22

7 180 178.02 = 175.02 + .10(205 – 175.02) 192.61

8 182 178.22 = 178.02 + .10(180 – 178.02) 186.30

9 ? 178.59 = 178.22 + .10(182 – 178.22) 184.15

69
 Impact of Different 

225 –

Actual  = .5
demand
200 –
Demand

175 –

 = .1
150 – | | | | | | | | |
1 2 3 4 5 6 7 8 9
Quarter
70
 Impact of Different 
225 –

Actual  = .5
► Chose
200 – high of 
values
demand
when underlying average
Demand

is likely to change
175 –
► Choose low values of 
when underlying average  = .1
is stable|
150 – | | | | | | | |
1 2 3 4 5 6 7 8 9
Quarter
71
 Choosing 
The objective is to obtain the most
accurate forecast no matter the
technique
We generally do this by selecting the
model that gives us the lowest forecast
error
Forecast error = Actual demand – Forecast value
= At – Ft

72
 Forecast Error Measurement
The forecast error measurement belongs to any one of the following

• to know the magnitude of error

• to get an idea on biasness of forecast

• to get an idea on revision of parameters

73
Common Measures of Error

Mean Absolute Deviation (MAD)

MAD =
å Actual - Forecast
n

where, n is the number deviations available

74
 Determining the MAD
ACTUAL
TONNAGE FORECAST WITH
QUARTER UNLOADED FORECAST WITH  = .10  = .50
1 180 175 175

2 168 175.50 = 175.00 + .10(180 – 175) 177.50

3 159 174.75 = 175.50 + .10(168 – 175.50) 172.75

4 175 173.18 = 174.75 + .10(159 – 174.75) 165.88

5 190 173.36 = 173.18 + .10(175 – 173.18) 170.44

6 205 175.02 = 173.36 + .10(190 – 173.36) 180.22

7 180 178.02 = 175.02 + .10(205 – 175.02) 192.61

8 182 178.22 = 178.02 + .10(180 – 178.02) 186.30

9 ? 178.59 = 178.22 + .10(182 – 178.22) 184.15

75
 Determining the MAD
ACTUAL FORECAST ABSOLUTE FORECAST ABSOLUTE
TONNAGE WITH DEVIATION WITH DEVIATION
QUARTER UNLOADED  = .10 FOR  = .10  = .50 FOR  = .50
1 180 175 5.00 175 5.00

2 168 175.50 7.50 177.50 9.50

3 159 174.75 15.75 172.75 13.75

4 175 173.18 1.82 165.88 9.12

5 190 173.36 16.64 170.44 19.56

6 205 175.02 29.98 180.22 24.78

7 180 178.02 1.98 192.61 12.61

8 182 178.22 3.78 186.30 4.30

Sum of absolute deviations: 82.45 98.62

Σ|Deviations|
MAD = 10.31 12.33
n
76
Common Measures of Error

Mean Squared Error (MSE)

2

MSE =
å (Forecast errors)
n

MSE penalise deviations with large magnitude

77
 Determining the MSE
ACTUAL
TONNAGE FORECAST FOR
QUARTER UNLOADED  = .10 (ERROR)2
1 180 175 52 = 25
2 168 175.50 (–7.5)2 = 56.25
3 159 174.75 (–15.75)2 = 248.06
4 175 173.18 (1.82)2 = 3.31
5 190 173.36 (16.64)2 = 276.89
6 205 175.02 (29.98)2 = 898.80
7 180 178.02 (1.98)2 = 3.92
8 182 178.22 (3.78)2 = 14.29
Sum of errors squared = 1,526.52

MSE =
å (Forecast errors)
= 1,526.52 / 8 = 190.8
n
78
Common Measures of Error

Mean Absolute Percent Error (MAPE)

n

å100 Actual -Forecast

i i
/ Actuali
i=1
MAPE =
n

79
 Determining the MAPE
ACTUAL
TONNAGE FORECAST FOR ABSOLUTE PERCENT ERROR
QUARTER UNLOADED  = .10 100(ERROR/ACTUAL)
1 180 175.00 100(5/180) = 2.78%
2 168 175.50 100(7.5/168) = 4.46%
3 159 174.75 100(15.75/159) = 9.90%
4 175 173.18 100(1.82/175) = 1.05%
5 190 173.36 100(16.64/190) = 8.76%
6 205 175.02 100(29.98/205) = 14.62%
7 180 178.02 100(1.98/180) = 1.10%
8 182 178.22 100(3.78/182) = 2.08%
Sum of % errors = 44.75%

MAPE =
å absolute percent error 44.75%
= = 5.59%
n 8
80
 Comparison of Forecast Error
Rounded Absolute Rounded Absolute
Actual Forecast Deviation Forecast Deviation
Tonnage with for with for
Quarter Unloaded  = .10  = .10  = .50  = .50
1 180 175 5.00 175 5.00
2 168 175.5 7.50 177.50 9.50
3 159 174.75 15.75 172.75 13.75
4 175 173.18 1.82 165.88 9.12
5 190 173.36 16.64 170.44 19.56
6 205 175.02 29.98 180.22 24.78
7 180 178.02 1.98 192.61 12.61
8 182 178.22 3.78 186.30 4.30
82.45 98.62

81
 Comparison of Forecast Error
∑ |deviations|
Rounded Absolute Rounded Absolute
Actual Forecast Deviation Forecast Deviation
MAD =
Tonnage with for with for
Quarter Unloaded
n
a = .10 a = .10  = .50  = .50
1 For  180
= .10 175 5.00 175 5.00
2 168 175.5 7.50 177.50 9.50
3 159 = 82.45/8
174.75 = 10.31
15.75 172.75 13.75
4 175 173.18 1.82 165.88 9.12
5 For  190
= .50 173.36 16.64 170.44 19.56
6 205 = 98.62/8
175.02 = 29.98
12.33 180.22 24.78
7 180 178.02 1.98 192.61 12.61
8 182 178.22 3.78 186.30 4.30
82.45 98.62

82
 Comparison of Forecast Error
∑ (forecast errors)
Rounded
2
Absolute Rounded Absolute
MSE =Tonnage
Actual Forecast Deviation Forecast Deviation

Quarter Unloaded
n
with
a = .10
for
a = .10
with
 = .50
for
 = .50
1 For  180
= .10 175 5.00 175 5.00
2 168 175.5 7.50 177.50 9.50
3 = 1,526.54/8
159 174.75 = 190.82
15.75 172.75 13.75
4 175 173.18 1.82 165.88 9.12
5 For  190
= .50 173.36 16.64 170.44 19.56
6 205 175.02
= 1,561.91/8 = 29.98
195.24 180.22 24.78
7 180 178.02 1.98 192.61 12.61
8 182 178.22 3.78 186.30 4.30
82.45 98.62
MAD 10.31 12.33

83
 Comparison of Forecast Error
n
∑100|deviation
Rounded i|/actualRounded
Absolute i Absolute
=Actuali = 1
MAPE Tonnage Forecast
with
Deviation
for
Forecast
with
Deviation
for
Quarter Unloaded a = .10 n a = .10 a = .50  = .50
1 For 
180= .10 175 5.00 175 5.00
2 168 175.5 7.50 177.50 9.50
3 159 = 44.75/8
174.75 =15.75
5.59% 172.75 13.75
4 175 173.18 1.82 165.88 9.12
5 For 
190= .50 173.36 16.64 170.44 19.56
6 205 175.02
= 54.05/8 =29.98
6.76% 180.22 24.78
7 180 178.02 1.98 192.61 12.61
8 182 178.22 3.78 186.30 4.30
82.45 98.62
MAD 10.31 12.33
MSE 190.82 195.24

84
 Comparison of Forecast Error
Rounded Absolute Rounded Absolute
Actual Forecast Deviation Forecast Deviation
Tonnage with for with for
Quarter Unloaded  = .10  = .10  = .50  = .50
1 180 175 5.00 175 5.00
2 168 175.5 7.50 177.50 9.50
3 159 174.75 15.75 172.75 13.75
4 175 173.18 1.82 165.88 9.12
5 190 173.36 16.64 170.44 19.56
6 205 175.02 29.98 180.22 24.78
7 180 178.02 1.98 192.61 12.61
8 182 178.22 3.78 186.30 4.30
82.45 98.62
MAD 10.31 12.33
MSE 190.82 195.24
MAPE 5.59% 6.76%
85
Values of Dependent Variable (y-values) Least Squares Method
Actual observation Deviation7
(y-value)

Deviation5 Deviation6

Deviation3
Least squares method minimizes the
sum of Deviation
the squared
4
errors (deviations)

Deviation1
(error) Deviation2
Trend line, y^ = a + bx

| | | | | | |
1 2 3 4 5 6 7
Figure 4.4
Time period
86
 Least Squares Method
Equations to calculate the regression variables

ŷ = a + bx

b=
å xy - nxy
å x - nx
2 2

a = y - bx
87
 Least Squares Example

ELECTRICAL ELECTRICAL
YEAR POWER DEMAND YEAR POWER DEMAND
1 74 5 105
2 79 6 142
3 80 7 122
4 90

88
 Least Squares Example
ELECTRICAL POWER
YEAR (x) DEMAND (y) x2 xy
1 74 1 74
2 79 4 158
3 80 9 240
4 90 16 360
5 105 25 525
6 142 36 852
7 122 49 854
Σx = 28 Σy = 692 Σx2 = 140 Σxy = 3,063

x=
å x 28
= =4 y=
å y 692
= = 98.86
n 7 n 7

89
 Least Squares Example
å xy - nxy 3,063 - ( 7) ( 4) (98.86) 295
b= = POWER
ELECTRICAL = = 10.54
YEAR (x) å x - nxDEMAND (y)140 - (7) ( 4 ) x 28
2 2 2 2 xy
1 74 1 74
2 79 4 158
3
a = y - bx = 98.8680 ()
-10.54 4 = 56.70 9 240
4 90 16 360
5 Thus,
105 ŷ = 56.70 +10.54x25 525
6 142 36 852
7 122 49 854
Σx = 28 Σy = 692 Σx2 = 140 Σxy = 3,063

x=
å
Demandx in
= =4 y=
å
28year 8 = 56.70 y+ 10.54(8)
=
692
= 98.86
n 7 = 141.02,
n or 141
7 megawatts

90
 Least Squares Example
Trend line,
160 – y^ = 56.70 + 10.54x
150 –
Power demand (megawatts)

140 –
130 –
120 –
110 –
100 –
90 –
80 –
70 –
60 –
50 –
| | | | | | | | |
1 2 3 4 5 6 7 8 9
Year Figure 4.5
91
Least Squares Requirements

1. We always plot the data to insure a

linear relationship
2. We do not predict time periods far
beyond the database
3. Deviations around the least squares
line are assumed to be random

92
 Associative Forecasting

Used when changes in one or more independent

variables can be used to predict the changes in
the dependent variable

Most common technique is linear

regression analysis

We apply this technique just as we did

in the time-series example

93
 Associative Forecasting
Forecasting an outcome based on predictor
variables using the least squares technique

y^ = a + bx

where y^ = value of the dependent variable (in our example,

sales)
a = y-axis intercept
b = slope of the regression line
x = the independent variable

94
 Associative Forecasting Example
NODEL’S SALES AREA PAYROLL NODEL’S SALES AREA PAYROLL
(IN $ MILLIONS), y (IN $ BILLIONS), x (IN $ MILLIONS), y (IN $ BILLIONS), x
2.0 1 2.0 2
3.0 3 2.0 1
2.5 4 3.5 7

4.0 –
Nodel’s sales
(in$ millions)

3.0 –

2.0 –

1.0 –

| | | | | | |

0 1 2 3 4 5 6 7
Area payroll (in $ billions)

95
Associative Forecasting Example
SALES, y PAYROLL, x x2 xy
2.0 1 1 2.0
3.0 3 9 9.0
2.5 4 16 10.0
2.0 2 4 4.0
2.0 1 1 2.0
3.5 7 49 24.5
Σy = 15.0 Σx = 18 Σx2 = 80 Σxy = 51.5

x=
å x 18
= =3 y=
å y 15
= = 2.5
6 6 6 6

b=
å xy - nxy 51.5 - (6)(3)(2.5)
= = .25 a = y - bx = 2.5 - (.25)(3) = 1.75
å x - nx
2 2
80 - (6)(3 ) 2

96
Associative Forecasting Example
SALES, y PAYROLL, x x2 xy
2.0 1 1 2.0
3.0 3 9 9.0
ŷ = 1.75 + .25x
2.5 4 16 10.0
2.0 2 Sales = 1.75
4 + .25(payroll)
4.0
2.0 1 1 2.0
3.5 7 49 24.5
Σy = 15.0 Σx = 18 Σx2 = 80 Σxy = 51.5

x=
å x 18
= =3 y=
å y 15
= = 2.5
6 6 6 6

b=
å xy - nxy 51.5 - (6)(3)(2.5)
= = .25 a = y - bx = 2.5 - (.25)(3) = 1.75
å x - nx
2 2
80 - (6)(3 ) 2

97
Associative Forecasting Example

If payroll next year is estimated to be $6 billion,

then:

Sales (in $ millions) = 1.75 + .25(6)

= 1.75 + 1.5 = 3.25

Sales = $3,250,000

98
Associative Forecasting Example

If payroll4.0
next
– year is estimated to be $6 billion,
then: 3.25
Nodel’s sales
(in$ millions)

3.0 –

2.0 –
Sales (in$ millions) = 1.75 + .25(6)
1.0 –
= 1.75 + 1.5 = 3.25
| | | | | | |
0 1 2 3 4 5 6 7
Sales = $3,250,000
Area payroll (in $ billions)

99
 Standard Error of the Estimate
► A forecast is just a point estimate of a
future value
► This point is
actually the
4.0 –
mean of a 3.25
Nodel’s sales
probability (in$ millions)
3.0 –
Regression line,
distribution 2.0 – ŷ = 1.75 + .25x

1.0 –

| | | | | | |
0 1 2 3 4 5 6 7
Figure 4.9 Area payroll (in $ billions)
Standard Error of the Estimate

S y,x =
å ( y - y c
) 2

n-2

where y = y-value of each data point

yc = computed value of the dependent variable,
from the regression equation
n = number of data points
 Standard Error of the Estimate
Computationally, this equation is
considerably easier to use

S y,x =
å - aå y - bå xy
y 2

n-2

We use the standard error to set up

prediction intervals around the point
estimate
 Standard Error of the Estimate

S y,x =
å - aå y - bå xy
y 2

=
39.5 -1.75(15.0) - .25(51.5)
n-2 6-2
= .09375
= .306 (in $ millions)
4.0 –
Nodel’s sales 3.25
(in$ millions) 3.0 –

The standard error 2.0 –

of the estimate is
1.0 –
$306,000 in sales
| | | | | | |
0 1 2 3 4 5 6 7
Area payroll (in $ billions)
 Correlation

► How strong is the linear relationship

between the variables?
► Correlation does not necessarily imply
causality!
► Coefficient of correlation, r, measures
degree of association
► Values range from -1 to +1

104
 Correlation Coefficient

nå xy - å xå y
r=
é ùé
2 ù 2
êënå x -
2
( )
å x úûêënå y -
2
( )
å y úû

105
 Correlation Coefficient
y y

x x
(a) Perfect negative (e) Perfect positive
correlation y y correlation

y
x x
(b) Negative correlation (d) Positive correlation

x
(c) No correlation

High Moderate Low Low Moderate High

| | | | | | | | |

–1.0 –0.8 –0.6 –0.4 –0.2 0 0.2 0.4 0.6 0.8 1.0
Correlation coefficient values

106
 Correlation Coefficient
y x x2 xy y2
2.0 1 1 2.0 4.0
3.0 3 9 9.0 9.0
2.5 4 16 10.0 6.25
2.0 2 4 4.0 4.0
2.0 1 1 2.0 4.0
3.5 7 49 24.5 12.25
Σy = 15.0 Σx = 18 Σx2 = 80 Σxy = 51.5 Σy2 = 39.5

(6)(51.5) – (18)(15.0)
r=
é(6)(80) – (18) 2 ùé(16)(39.5) – (15.0) 2 ù
ë ûë û

309 - 270 39 39
= = = = .901
(156)(12) 1,872 43.3

107
 Correlation
► Coefficient of Determination, r2,
measures the percent of change in y
predicted by the change in x
► Values range from 0 to 1
► Easy to interpret

For the previous example:

r = .901
r2 = .81
108
 Monitoring and Controlling
Forecasts
Tracking Signal
► Measures how well the forecast is predicting
actual values
► Ratio of cumulative forecast errors to mean
absolute deviation (MAD)
► Good tracking signal has low values
► If forecasts are continually high or low, the
forecast has a bias error

109
 Monitoring and Controlling
Forecasts

Tracking Cumulative error

signal =
MAD

=
å (Actual demand in period i -Forecast demand in period i)
å Actual -Forecast
n

110
 Tracking Signal
Figure 4.11
Signal exceeding limit
Tracking signal
Upper control limit
+

0 MADs Acceptable
range

–
Lower control limit

Time

111
 Tracking Signal Example
(-4<=MAD<= 4)
ABSOLUTE CUM ABS TRACKING
ACTUAL FORECAST CUM FORECAST FORECAST SIGNAL (CUM
QTR DEMAND DEMAND ERROR ERROR ERROR ERROR MAD ERROR/MAD)
1 90 100

2 95 100

3 115 100

4 100 110

5 125 110

6 140 110

112
 Tracking Signal Example
(-4<=MAD<= 4)
ABSOLUTE CUM ABS TRACKING
ACTUAL FORECAST CUM FORECAST FORECAST SIGNAL (CUM
QTR DEMAND DEMAND ERROR ERROR ERROR ERROR MAD ERROR/MAD)
1 90 100 –10 –10 10 10 10.0 –10/10 = –1

2 95 100 –5 –15 5 15 7.5 –15/7.5 = –2

3 115 100 +15 0 15 30 10. 0/10 = 0

4 100 110 –10 –10 10 40 10. 10/10 = –1

5 125 110 +15 +5 15 55 11.0 +5/11 = +0.5

6 140 110 +30 +35 30 85 14.2 +35/14.2 = +2.5

At the end of quarter 6, MAD =

å Forecast errors 85
= = 14.2
n 6
Cumulative error 35
Tracking signal = = = 2.5 MADs
MAD 14.2
113
• How Firms has to decide what the upper and lower tracking limits should be?
• Limit should not so low as to be triggered with every small forecast error and not
so high as to allow bad forecasts to be regularly overlooked.
1 MAD = ± 0.8 std(approximately)
2 MAD = ± 1.6 std
3 MAD = ± 2.4 std
4 MAD = ± 3.2 std

114
 Quiz
1) Forecasts used for new product planning, capital expenditures, facility
location or expansion, and R&D typically utilize a:
A) short-range time horizon.
B) medium-range time horizon.
C) long-range time horizon.
D) naive method, because there is no data history.
E) trend extrapolation.

2) The three major types of forecasts used by organizations in planning future

operations are:
A) strategic, tactical, and operational.
B) economic, technological, and demand.
C) exponential smoothing, Delphi, and regression.
D) causal, time-series, and seasonal.
E) departmental, organizational, and territorial.

115
3) Which of the following uses three types of participants: decision makers,
staff personnel, and respondents?
A) jury of executive opinion
B) sales force composite
C) Delphi method
D) associative models
E) time series

4) Which of the following techniques uses variables such as price and

promotional expenditures, which are related to product demand, to predict
demand?
A) associative models
B) exponential smoothing
C) weighted moving average
D) moving average
E) trend projection

116
5. What is the difference between an associative model and a time-series
model?
A time-series model uses only historical values of the quantity of interest to
predict future values of that quantity.
The associative model, on the other hand, incorporates the variables or factors
that might influence the quantity being forecast.

6) Which of the following statements about time-series forecasting is true?

A) It is always based on the assumption that future demand will be the same as
past demand.
B) It makes extensive use of the data collected in the qualitative approach.
C) It is based on the assumption that the analysis of past demand helps predict
future demand.
D) Because it accounts for trends, cycles, and seasonal patterns, it is always
more powerful than associative forecasting.

117
7) Which of the following statements about time-series forecasting is true?
A) It is always based on the assumption that future demand will be the same as
past demand.
B) It makes extensive use of the data collected in the qualitative approach.
C) It is based on the assumption that the analysis of past demand helps predict
future demand.
D) Because it accounts for trends, cycles, and seasonal patterns, it is always
more powerful than associative forecasting.

8) Increasing the number of periods in a moving average will accomplish

greater smoothing, but at the expense of:
A) manager understanding.
B) accuracy.
C) stability.
D) sensitivity to real changes in the data.
E) All of the above are diminished when the number of periods increases.

118
23)
9. Which of the following statements comparing exponential smoothing to the
weighted moving average technique is TRUE?
A) Exponential smoothing is more easily used in combination with the Delphi
method.
B) More emphasis can be placed on recent values using the weighted moving
average.
C) Exponential smoothing is considerably more difficult to implement on a
computer.
D) Exponential smoothing typically requires less record keeping of past data.
E) Exponential smoothing allows one to develop forecasts for multiple
periods, whereas the weighted moving average technique does not.

10) Which of the following is NOT a characteristic of exponential smoothing?

A) smoothes random variations in the data
B) uses an easily altered weighting scheme
C) weights each historical value equally
D) has minimal data storage requirements
E) uses the previous period's forecast

119
11) Which of the following smoothing constants would make an exponential
smoothing forecast equivalent to a naive forecast?
A) 0
B) 1 divided by the number of periods
C) 0.5
D) 1.0
E) cannot be determined

12)Which of the following values of alpha would cause exponential smoothing

to respond the SLOWEST to forecast errors?
A) 0.10
B) 0.2246
C) 0.50
D) 0.90
E) cannot be determined

120
13) For a given product demand, the time-series trend equation is 53 - 4x. The
negative sign on the slope of the equation:
A) is a mathematical impossibility.
B) is an indication that the forecast is biased, with forecast values lower than actual
values.
C) is an indication that product demand is declining.
D) implies that the coefficient of determination will also be negative.
E) implies that the cumulative error will be negative.

14) If Brandon Edward were working to develop a forecast using a moving

averages approach, but he noticed a detectable trend in the historical data, he
should:
A) use weights to place more emphasis on recent data.
B) use weights to minimize the importance of the trend.
C) change to an associative multiple regression approach.
D) use a simple moving average.
E) change to a qualitative approach.

15) A measure of forecast error that does not depend upon the magnitude of the
item being forecast is the ________.

121
16) Distinguish between a weighted moving average model and an exponential
smoothing model.
Exponential smoothing is a weighted moving average model wherein previous
values are weighted in a specific manner--in particular, all previous values are
weighted with a set of weights that decline exponentially. Also, exponential
smoothing involves little record keeping of past data.

17)What does it mean to "decompose" a time series?

Answer: To decompose a time series means to break past data down into
components of trend, seasonality, cycles, and random variations, and then to
project them forward

122