Forecasting: Types of forecasts; Uses of

forecasts; forecasting techniques; trend

line; market survey; exponential
smoothing; correlation analysis; Historical
estimate; Seasonability distribution;
Forecast control and Monitoring.
Problems.

Hari Dhakal
What is Forecasting?

The art and science of predicting

future events.

“Prediction is very difficult and not quantifiable

especially if it's about the future, so finding an
engineered way to reduce error is the only solution.
So Forecasting has been engineered in quantifiable
manner by minimizing the errors in every possible
way”.
What is forecasting all about?

Demand for a Product We try to predict the

future by looking back
at the past

Predicted
demand
looking
Time back six
Jan Feb Mar Apr May Jun Jul Aug months
Actual demand (past sales)
Predicted demand
Past Data Current Data

Forecast Forecast
Generation Control

Managerial
Judgement
&
Experience

Modified
Forecast
Forecasting System
 Forecasting
 Objective
 Scientific
 Free from ‘BIAS’
 Reproducible
 Error Analysis Possible

Prediction
 Subjective
 Intuitive
 Individual BIAS
 Non - Reproducible
 Error Analysis Limited
Issues in Forecasting

1. A forecast is only as good as the information included in the

forecast (past data)
2. History is not a perfect predictor of the future (i.e.: there is
no such thing as a perfect forecast)

Note : Forecasting is based on the assumption that the

past predicts the future! When forecasting, think
carefully whether or not the past is strongly related to
what you expect to see in the future.
 Uses of Forecasts

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

 Production Manager and Forecast
 A production manager is interested in the future
of his industry, demand, production etc. so
forecasting can be helpful for him.
 If he has knowledge about future then
 He can take appropriate decision to make the good
field of operation.
 He has enough time to solve the problems and to
implement his new ideas which can be helpful for
improving his job functioning.
 He can take action when sufficient time is in hand to
implement the plan.
 The main duty of production manager of any
industry is to improve the production by earning
more and more profit along with production.
 In order to have an idea about the production,
manager should have sufficient knowledge about
forecasting.
Types of Forecasts by Time Horizon
Quantitative
• Short-range forecast methods
– Usually < 3 months
• Job scheduling, worker assignments
Detailed
• Medium-range forecast use of
system
– 3 months to 3 years
• Sales/production planning

• Long-range forecast
– > 3 years
Design
• New product planning
of system
Qualitative
Methods
 Forecasting

Forecasting is essential for a number of planning

decisions

Long Term Forecast Decisions

 Useful in planning for new products introduction,
capital expenditures, facilities or plant expansion
and research and development.
Medium Term Forecast Decisions
 Useful in sales planning, aggregate
production planning , manpower planning ,
inventory policy and budgeting.

Short Term Forecast Decisions

 Used for planning, purchasing, job
assignments, Scheduling of job orders and
production level.
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 level
Inventory level
Factory capacity
 Product Life Cycle
Introduction Growth Maturity Decline
Product design Forecasting Standardization Little product
and critical Fewer product differentiation
development Product and changes, more Cost
critical process reliability minor changes minimization
Frequent product Competitive Optimum Overcapacity
and process product capacity in the industry
design changes improvements and Increasing Prune line to
Short production options
Strategy/Issues

stability of eliminate
runs Increase capacity process items not
High production Shift toward Long production returning good
costs product focus runs margin
Limited models Enhance Product Reduce
Attention to distribution improvement and capacity
quality cost cutting
Types of Forecasting methods

 Qualitative methods  Quantitative methods

Rely on subjective Rely on data and

Opinions from one or analytical techniques.
more experts.

Note: Some firms use either of the One approach. But

in practice, a combination of both usually seems most
effective.
 Methods Of Forecasting

(a) Subjective/ Qualitative or intuitive

methods
 Jury of executive opinion
 Sales force composite
 Consumer Market Survey
 DELPHI method
 Qualitative method
Used when
 Not enough time to gather and analyze data to
forecast.
 Available data may be absolute, more up to date
information might not yet be available.
 When historical data is unavailable
 For new design
 Redesign of existing products
 Qualitative Methods
 Jury of executive opinion (Pool opinions of
high-level experts, sometimes augment by
statistical models)
 DELPHI method (Panel of experts, queried
iteratively until consensus is reached)
 Sales force composite (Estimates from
individual salespersons are reviewed for
reasonableness, then aggregated)
 Consumer Market Survey (Ask the customer)
 DELPHI
A structured method of obtaining responses
from experts.
 Utilizes the vast knowledge base of experts
 Eliminates subjective bias and
‘influencing’ by members through
anonymity
 Iterative in character with statistical
summary at end of each round (Generally
3 rounds)
 Consensus (or Divergent Viewpoints)
 usually emerge at the end of the
exercise.
 DELPHI
Expert 1

Coordinator Expert 2

Expert n

A Statistical
• Mean summary
•Median can be given
1990 1995 2000 2005 2010 •Std. deviation at end of
year each round
Example:
Analyzing the petroleum export of Nepal in certain years, and
forecasting the best forecast by experts opinion.
 DELPHI (Contd.)

Round
1
Moving
Towards
Consensus Round
2

Round
3
 DELPHI (Contd.)
Round
1
Moving
Towards
Divergent
View Points Round
2

Round
3
 Quantitative Approaches
 Used when situation is ‘stable’ and
historical data exist
 Existing products
 Current technology
 Involves mathematical techniques
 Quantitative Methods
1. Naive approach
2. Moving averages
time-series
3. Exponential models
smoothing
4. Trend projection
associative
5. Linear regression model

Time Series Model : It predicts on the assumption that the future

is a function of the past.
Associative Model : incorporate the variables or factors that
might influence the quantity being forecast.
 Time Series Forecasting

 Set of evenly spaced numerical data

 Obtained by observing response variable at
regular time periods (weekly, monthly yearly
etc.)
 Forecast based only on past values, no other
variables important
 Assumes that factors influencing past and
present will continue influence in future
 Time Series Component
Which are commonly observed
demand patterns
 Trend
 Cyclic
 Seasonal
 Random
 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)
Product Demand charted over 4 years with a Growth Trend and
Seasonality Indicated (Eg: Nokia cell phones)
 Trend Component
 Persistent, overall
upward or
Linear
downward pattern Trend
 Changes due to
population, D

technology, age,
culture, etc.
 Typically several t

years duration
 Seasonal Component
 Regular pattern of up and down
fluctuations
 Due to weather, customs, etc.
 Occurs within a single year
D Seasonal
Pattern with
Growth

t
 Cyclical Component
 Repeating up and down movements
 Affected by business cycle, political, and
economic factors
 Multiple years duration

 0 5 10 15 20
 Random Component
 Erratic, unsystematic, ‘residual’ fluctuations
 Due to random variation or unforeseen events
 Short duration
and nonrepeating

 M T W T
F
 Naive Approach
 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 for a Product.
 Sometimes cost effective and efficient.
 Can be good starting point.
 Moving Average Method

 Forecast based on an average of recent

values.
 It is a series of arithmetic means.
 Used if little or no trend.
 Used often for smoothing.
 Provides overall impression of data over time.

∑ demand in previous n periods

Moving average =
n
Ramesh Automotive workshop wants a 3 months moving average
forecast, for Yamaha sparkplug sales in his workshop.

Actual 3-Month
Month Sparkplug Sales Moving Average
January 10
February 12
March 13
(10+ 12 + 13)/3 = 11 2/3
April 16
(12 + 13 + 16)/3 = 13 2/3
May 19
(13 + 16 + 19)/3 = 16
June 23
(16 + 19 + 23)/3 = 19 1/3
July 26
 Graph of Moving Average
Moving
Average
30 – Forecast
28 –
Actual Sales
26 –
24 –
Shed Sales

22 –
20 –
18 –
16 –
14 –
12 –
10 –
| | | | | | | | | | | |
J F M A M J J A S O N D
 Characteristics Of Moving
Averages

Dt Dt

t t

Actual Demand

Moving Average Demand

(1) MOVING AVERAGES LAG A TREND

 Dt

(2) MOVING AVERAGES ARE OUT OF PHASE

FOR CYCLIC DEMAND
Dt

(3) MOVING AVERAGES FLATTEN PEAKS

 Weighted Moving Average
 Forecast based on a moving average with
weights that vary.
 Used when some trend might be present
 Older data usually less important
 Weights based on experience and intuition
∑ (weight for period n)
Weighted x (demand in period n)
moving average =

∑ weights
 Weighted Moving Average
Ramesh Automotive workshop wants a 3 months weighted moving
average forecast, for Yamaha sparkplug sales in his workshop.
Weights Applied Period
3 Last month
2 Two months ago
1 Three months ago
6 Sum of weights

Actual 3-Month Weighted

Month Sparkplug Sales Moving Average

January 10
February 12
March 13
April 16 [(3 x 13) + (2 x 12) + (10)]/6 = 121/6
May 19 [(3 x 16) + (2 x 13) + (12)]/6 = 141/3
June 23 [(3 x 19) + (2 x 16) + (13)]/6 = 17
July 26 [(3 x 23) + (2 x 19) + (16)]/6 = 201/2
 Moving Average And
Weighted Moving Average
Weighted
30 – moving
average
25 –
Sales demand

20 – Actual
sales
15 –
Moving
10 – average

5 –
| | | | | | | | | | | |
J F M A M J J A S O N D
Moving Average
Advantage:
Easy to compute and easy to understand.
Disadvantage:
All values in the average are weighted
equally. That’s why lag in trend, flattening peaks happen
in demands.

Weighted Moving Average:

Similar to moving average.
It assigns more weight to the most recent
values in a time series.

Note: most recent observations must be better indicators

of the future than older observations.
 Exponential Smoothing
 Form of moving average with weights
following an exponential distribution
 Weights decline exponentially
 Most recent data weighted most
 Requires smoothing constant (Alpha)
 Ranges from 0 to 1
 Subjectively chosen
 Involves little record keeping of past data
 Exponential Smoothing
Forecast error =Actual Demand – Forecasted Demand
= (At - 1) - (Ft - 1 )

Ft  Ft 1   ( At 1  Ft 1 )
Forecast today=Forecast yesterday+(alpha)*(Forecast error yesterday)
Each new forecast is equal to the previous forecast plus a percentage of
the previous error.
Today’s forecast
Depends on yesterday’s (time-wise dependence, strong memory)
But it has to be corrected by forecast error only.
Therefore, we should give more weight to the more recent time
periods when forecasting.
Alpha = smoothing constant = percentage of the forecast error.
 Exponential Smoothing
New forecast = Last period’s forecast
+  (Last period’s actual demand
– Last period’s forecast)
Ft  Ft 1   ( At 1  Ft 1 )
Where Ft = new forecast
Ft – 1 = previous forecasted demand
At – 1= previous actual demand
 = smoothing (or weighting)
constant (0 ≤  ≤ 1)
Ft =  At-1 +(1 - ) Ft-1
=  At-1 +(1 - ) [ At-2 +(1 - )2 Ft-2 ]
=……..
=  [At-1 +(1 - ) At-2 +(1 - )2 At-3 + …..
+ (1 - )t-1 At + (1 - )t F0]


 (1- )
 (1- )2
t-2 t-1 t

Weightages given to past data decline exponentially.

 Exponential Smoothing Example
 In January, a car dealer predicted February
demand for 142 Ford mustang. Actual
February demand was 153, Using a smoothing
constant chosen by management of  = 0.20,
the dealer wants to forecast March demand
using the exponential smoothing model.
Solution:
Predicted demand Ft – 1 = 142 Ford Mustangs
Actual demand At – 1 = 153
Smoothing constant  = .20
Exponential Smoothing Example

Predicted demand Ft – 1 = 142 Ford Mustangs

Actual demand At – 1 = 153
Smoothing constant  = .20

Ft  Ft 1   ( At 1  Ft 1 )
New forecast Ft = 142 + .2(153 – 142)
= 142 + 2.2
= 144.2 ≈ 144 cars
 Impact of Different 
 Choose high values of  when underlying
average is likely to change
 Choose low values of  when underlying
average is stable
225 –

Actual  = .5
demand
200 –
Demand

175 –

 = .1
150 – | | | | | | | | |
1 2 3 4 5 6 7 8 9
Quarter
 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
 Common Measures of Error
 MAD is computed by taking the sum of the absolute values of the
individual errors and dividing it by the number of periods of
Data(n).
Mean Absolute Deviation (MAD)

∑ |Actual - Forecast|
MAD = n

 Another way of measuring overall error, MSE is the average of

the squared differences between forecasted and observed
values.
Mean Squared Error (MSE)

∑ (Forecast Errors)2
MSE = n
 Common Measures of Error
 A problem with both MAD and MSE values depend on
the magnitude of the forecast.
 If the forecast items are measured in thousands then
simultaneously the values become large.
 To avoid that issue, MAPE is computed as the average of
the absolute differences between the forecasted and
actual values, expressed as a percentage of the actual
values.
Mean Absolute Percent Error (MAPE)
n
∑100(|Actuali - Forecasti|/Actuali)
MAPE = i=1
n
 Example
 During the past 8 quarters, the Inventory
Store of Jagadamba steels has unloaded large
quantities of Iron ore from India as given in
the table. The Operations Manager wants to
test the forecasting method exponential
smoothing to see how well this method
works in predicting tonnage unloaded.
 He guesses that the forecast of Iron ore
unloaded in the first quarter was 175
tons.Two values of  are to be examined  =
.10 and  = .50.
a. Determine MAD, recommend the best.
b. Compute MSE for  = .10.
c. Calculate MAPE when  = .10.
 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
Apply, Ft  Ft 1   ( At 1  Ft 1 )
175.5 = 175 + .10 (180- 175)
 Comparison of Forecast Error
∑ |deviations|
Rounded Absolute Rounded Absolute
MADActual
= Forecast Deviation Forecast Deviation
Tonnage n
with for with for
Quarter Unloaded a = .10 a = .10  = .50  = .50
1
For  180
= .10 175 5.00 175 5.00
2 168 = 82.45/8
175.5 = 10.31
7.50 177.50 9.50
3 159 174.75 15.75 172.75 13.75
4 For  175
= .50 173.18 1.82 165.88 9.12
5 190 173.36 16.64 170.44 19.56
6 205 = 98.62/8
175.02 = 12.33
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
 Comparison of Forecast Error
∑ (forecast errors)2
Rounded Absolute Rounded Absolute
MSE = Actual
Tonnage
n
Forecast
with
Deviation
for
Forecast
with
Deviation
for
Quarter Unloaded a = .10 a = .10  = .50  = .50
1
For  = .10
180 175 5.00 175 5.00
2 = 1,526.54/8
168 175.5 = 190.82
7.50 177.50 9.50
3 159 174.75 15.75 172.75 13.75
4 For  175
= .50 173.18 1.82 165.88 9.12
5 190 173.36 16.64 170.44 19.56
6 = 1,561.91/8
205 175.02 = 195.24
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

Note : On the basis of this comparison of the two MADs, a

smoothing constant of  = .10 is preferred to  = .50 because
its MAD is smaller.
 Comparison
n
of Forecast Error
∑100(|deviationi|/actuali )
Rounded Absolute Rounded Absolute
MAPE =Actual
i=1
Forecast Deviation Forecast Deviation
Tonnage with n for with for
Quarter Unloaded a = .10 a = .10 a = .50  = .50
1
For 180= .10 175 5.00 175 5.00
2 168 = 44.75/8
175.5 = 5.59%
7.50 177.50 9.50
3 159 174.75 15.75 172.75 13.75
4 For 175= .50 173.18 1.82 165.88 9.12
5 190 173.36 16.64 170.44 19.56
6 205 = 54.05/8
175.02 = 6.76%
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
 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%
Conclusion : On the basis of comparison of MAD,MSE and MAPE at  =
.10 and  = .50. Smoothing constant at  = .10 must be preferred because of
the deviation, errors and percent errors are less compared to the other.
 Exponential Smoothing with Trend
Adjustment
When a trend is present, exponential smoothing
must be modified to respond to trend
Forecast Exponentially Exponentially
including (FITt) = smoothed (Ft) + smoothed (Tt)
trend forecast trend

Ft = (Actual Demand last period) + (1 - )(Forecast last period

+ Trend estimate last period)
Ft = (At - 1) + (1 - )(Ft - 1 + Tt - 1)
Tt = (Forecast this period - Forecast last period) + (1 - )(Trend
estimate last period)
Tt = (Ft - Ft - 1) + (1 - )Tt - 1
Exponential Smoothing with Trend
Adjustment
Ft = exponentially smoothed forecast of the data
series in period t
Tt= exponentially smoothed trend in period t
At= Actual demand in period t
 = smoothing constant for the average
 = smoothing constant for the trend

Step 1: Compute Ft
Step 2: Compute Tt
Step 3: Calculate the forecast FITt = Ft + Tt
Example

• A car manufacturer Aston Martin wants to forecast

the demand for a vanquish model.
• Past data shows that there is an increasing trend.
• The company assumes the initial forecast for month 1
was 11 units and the trend over that period was 2
units.
• Smoothing constants are α = 0.2 β =0.4
 Exponential Smoothing with Trend
Adjustment Example
A trend adjusted exponential smoothing model is employed
Forecast
Actual Smoothed Smoothed Including
Month(t) Demand (At) Forecast, Ft Trend, Tt Trend, FITt
1 12 11 2 13.00
2 17
3 20
4 19
5 24
6 21
7 31
8 28
9 36
10 ?
Exponential Smoothing with Trend
Adjustment Example
Forecast
Actual Smoothed Smoothed Including
Month(t) Demand (At) Forecast, Ft Trend, Tt Trend, FITt
1 12 11 2 13.00
2 17
3 20
4 19
5 24 Step 1: Forecast for Month 2
6 21
Ft = (At - 1) + (1 - )(Ft - 1 + Tt - 1)
7 31
F2 =  A1 + (1 - )(F1 + T1)
8 28
9 36 F2 = (.2)(12) + (1 - .2)(11 + 2)
10 = 2.4 + 10.4 = 12.8 units
Exponential Smoothing with Trend
Adjustment Example
Forecast
Actual Smoothed Smoothed Including
Month(t) Demand (At) Forecast, Ft Trend, Tt Trend, FITt
1 12 11 2 13.00
2 17 12.80
3 20
4 19
5 24 Step 2: Trend for Month 2
6 21
T2 = (F2 - F1) + (1 - )T1
7 31
8 28 T2 = (.4)(12.8 - 11) + (1 - .4)(2)
9 36 = .72 + 1.2 = 1.92 units
10
Exponential Smoothing with Trend
Adjustment Example
Forecast
Actual Smoothed Smoothed Including
Month(t) Demand (At) Forecast, Ft Trend, Tt Trend, FITt
1 12 11 2 13.00
2 17 12.80 1.92
3 20
4 19
5 24 Step 3: Calculate FIT for Month 2
6 21
FIT2 = F2 + T2
7 31
FIT2 = 12.8 + 1.92
8 28
9 36 = 14.72 units
10
Exponential Smoothing with Trend
Adjustment Example
Forecast
Actual Smoothed Smoothed Including
Month(t) Demand (At) Forecast, Ft Trend, Tt Trend, FITt
1 12 11 2 13.00
2 17 12.80 1.92 14.72
3 20 15.18 2.10 17.28
4 19 17.82 2.32 20.14
5 24 19.91 2.23 22.14
6 21 22.51 2.38 24.89
7 31 24.11 2.07 26.18
8 28 27.14 2.45 29.59
9 36 29.28 2.32 31.60
10 32.48 2.68 35.16
Exponential Smoothing with Trend
Adjustment Example
35 –

30 – Actual demand (At)

Product demand

25 –

20 –

15 –

10 – Forecast including trend (FITt)

with  = .2 and  = .4
5 –

0 – | | | | | | | | |
1 2 3 4 5 6 7 8 9
Time (month)
 Trend Projections
 The last time series forecasting method.
 This technique fits a trend line to historical data
points to project into the medium to long-range
forecasts.
 Linear trends can be found using the least
squares technique.
y^ = a + bx
where ^y = computed value of the variable to be
predicted (dependent variable)
a = y-axis intercept
b = slope of the regression line(or the rate of
change in y for a given changes in x)
x = the independent variable
 Least Squares Method

Actual observation
Values of Dependent Variable

Deviation7
(y-value)

Deviation5 Deviation6

Deviation3

Deviation4

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

Time period
 Least Squares Method

Actual observation
Values of Dependent Variable

Deviation7
(y-value)

Deviation5 Deviation6

Deviation3
Least squares method minimizes the sum
of the squared errors (deviations) of the
vertical differences
Deviation4

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

Time period
 Least Squares Method
Equations to calculate the regression variables
^
y = a + bx
Sxy - nxy
b=
Sx2 - nx2
a = y - bx
Where,
b = slope of regression line
S  Summation sign
x = known values of the independent variable
y = known values of the dependent variable
x = average of the x- values
y = average of the y- values
n = number of data points or observations
 Least Squares Example
 A demand for electric power in Dhulikhel over the period 2011
to 2017 is shown in the following table, in megawatts. Nepal
Electricity Authority wants to forecast 2018 2019 demand by fitting a
straight line trend to these data.
Electrical Power
Year Demand (megawatt)
2011 74
2012 79
2013 80
2014 90
2015 105
2016 142
2017 122
 Time Electrical Power
Year Period (x) Demand (megawatt)(y) x2 xy
2011 1 74
2012 2 79
2013 3 80
2014 4 90
2015 5 105
2016 6 142
2017 7 122
∑x = 28 ∑y = 692 ∑x2 = ∑xy =
n=7 x=4 y = 98.86

∑xy - nxy
b= =
∑x2 - nx2

a = y - bx =
 Time Electrical Power
Year Period (x) Demand (megawatt)(y) x2 xy
2011 1 74 1 74
2012 2 79 4 158
2013 3 80 9 240
2014 4 90 16 360
2015 5 105 25 525
2016 6 142 36 852
2017 7 122 49 854
∑x = 28 ∑y = 692 ∑x2 = 140 ∑xy = 3,063
x=4 y = 98.86

∑xy - nxy = 3,063 - (7)(4)(98.86)

b= =
2
∑x - nx 2 140 - (7)(42)
= 10.54
a = y - bx = 98.86 - 10.54(4) = 56.70
 Least Squares Example
Time Electrical Power
Year Period (x) Demand x2 xy
2011 1 74 1 74
2012 2 79 4 158
2013The trend
3 line is 80 9 240
2014 4 90 16 360
^
2015 y =5 56.70 + 10.54x
105 25 525
2016 6 142 36 852
2017 7 122 49 854
∑x = 28 ∑y = 692 ∑x2 = 140 ∑xy = 3,063
x=4 y = 98.86

∑xy - nxy 3,063 - (7)(4)(98.86)

b= = = 10.54
2
∑x - nx 2 140 - (7)(42)

a = y - bx = 98.86 - 10.54(4) = 56.70

 Least Squares Example
Trend line,
160 – y^= 56.70 + 10.54x
150 –
140 –
Power demand

130 –
120 –
110 –
100 –
90 –
80 –
70 –
60 –
50 –
| | | | | | | | |
2011 2012 2013 2014 2015 2016 2017 2018 2019
Year
 Least Squares Requirements
 We always plot the data to insure a linear
relationship.
 We do not predict time periods far beyond
the database.
 Deviations around the least squares line
are assumed to be random and normally
distributed.
 Seasonal Variations In Data
Steps in the process:
1. Find average historical demand for each season
2. Compute the average demand over all seasons
3. Compute a seasonal index for each season
4. Estimate next year’s total demand
5. Divide this estimate of total demand by the number of
seasons, then multiply it by the seasonal index for that
season
 Seasonal Index Example
Chaudhary Group- Electronics wants to develop monthly indices
for sales of LG Air Conditioners. Data from 2010 -2012 , by
months are available. Give decisions as Operations manager, and
calculate how the demand would be in different seasons.
Demand
Month 2010 2011 2012
Jan 80 85 105
Feb 70 85 85
Mar 80 93 82
Apr 90 95 115
May 113 125 131
Jun 110 115 120
Jul 100 102 113
Aug 88 102 110
Sept 85 90 95
Oct 77 78 85
Nov 75 72 83
Dec 82 78 80
Chaudhary Group- Electronics division wants to develop monthly indices for
sales of LG Air Conditioners. Data from 2010 -2012 , by months are available.
Calculate and Give decisions as Operations manager for how the demand would
be in different seasons.
Demand Average Average Seasonal
Month 2010 2011 2012 2010-2012 Monthly Index
Jan 80 85 105
Feb 70 85 85
Mar 80 93 82
Apr 90 95 115
May 113 125 131
Jun 110 115 120
Jul 100 102 113
Aug 88 102 110
Sept 85 90 95
Oct 77 78 85
Nov 75 72 83
Dec 82 78 80
Total Average annual demand = 1,128
Average monthly demand = 1,128/12 months = 94
Chaudhary Group- Electronics division wants to develop monthly indices for
sales of LG Air Conditioners. Data from 2010 -2012 , by months are available.
Calculate and Give decisions as Operations manager for how the demand would
be in different seasons.
Demand Average Average Seasonal
Month 2010 2011 2012 2010-2012 Monthly Index
Jan 80 85 105 90 94
Feb 70 85 85 80 94
Mar 80 93 82 85 94
Apr 90 95 115 100 94
May 113 125 131 123 94
Jun 110 115 120 115 94
Jul 100 102 113 105 94
Aug 88 102 110 100 94
Sept 85 90 95 90 94
Oct 77 78 85 80 94
Nov 75 72 83 80 94
Dec 82 78 80 80 94
Total Average annual demand = 1,128
Average monthly demand = 1,128/12 months = 94
 Average 2010-2012 monthly demand
Seasonal index =
Average monthly demand
= 90/94 = .957
Demand Average Average Seasonal
Month 2010 2011 2012 2010-2012 Monthly Index
Jan 80 85 105 90 94 0.957
Feb 70 85 85 80 94
Mar 80 93 82 85 94
Apr 90 95 115 100 94
May 113 125 131 123 94
Jun 110 115 120 115 94
Jul 100 102 113 105 94
Aug 88 102 110 100 94
Sept 85 90 95 90 94
Oct 77 78 85 80 94
Nov 75 72 83 80 94
Dec 82 78 80 80 94
 Demand Average Average Seasonal
Month 2010 2011 2012 2010-2012 Monthly Index
Jan 80 85 105 90 94 0.957
Feb 70 85 85 80 94 0.851
Mar 80 93 82 85 94 0.904
Apr 90 95 115 100 94 1.064
May 113 125 131 123 94 1.309
Jun 110 115 120 115 94 1.223
Jul 100 102 113 105 94 1.117
Aug 88 102 110 100 94 1.064
Sept 85 90 95 90 94 0.957
Oct 77 78 85 80 94 0.851
Nov 75 72 83 80 94 0.851
Dec 82 78 80 80 94 0.851
 If we expect the 2013 annual demand for LG Air Conditioners to be
1,200 units, We would use these seasonal indices to forecast the
monthly demand.
Demand Average Average Seasonal Seasonal
Month 2010 2011 2012 2010-2012 Demand
2013
Monthly Index
Jan 80 85 105 90 94 0.957 96
Feb 70 85 85 Forecast
80 for 2013 94 0.851 85
Mar 80 93 82 85 94 0.904 90
Apr 90 95 Expected
115 Annual Demand
100 = 1,200 94 1.064 106
May 113 125 131 123 94 1.309 131
Jun 110 115 120 1,200 115 94 1.223 122
Jul 100 102 Jan113 12 105x .957 = 96 94 1.117 112
Aug 88 102 110 100 94 1.064 106
Sept 85 90 Feb 95 1,200 90
x .851 = 85 94 0.957 96
Oct 77 78 85 12 80 94 0.851 85
Nov 75 72 83 80 94 0.851 85
Dec 82 78 80 80 94 0.851 85
 Seasonal Index Example
2013 Forecast
140 – 2012 Demand
130 – 2011 Demand
2010 Demand
120 –
Demand

110 –
100 –
90 –
80 –
70 –
| | | | | | | | | | | |
J F M A M J J A S O N D
Time
 Associative Forecasting
 Used when changes in one or more independent
variables can be used to predict the changes in
the dependent variable.

 Most common associative forecasting technique

is linear regression analysis.

 We apply this technique just as we did in the

time series example.
 Associative Forecasting
Forecasting an outcome based on predictor
variables using the least squares technique
^
y = a + bx
where ^
y = computed value of the variable to be predicted
(dependent variable)
a = y-axis intercept
b = slope of the regression line
x = the independent variable thought to predict the
value of the dependent variable
 Associative Forecasting Example
DBCC Construction Company renovates old homes in New
york. Over time, the company has found that its dollar volume
(stock or exchange-traded fund's share price times its
average volume) of renovation work is dependent on New york
area payroll. Management wants to establish a mathematical
relationship to help predict sales.
Sales Area Payroll
($ millions), y ($ billions), x
4.0 –

Sales ($ millions), y
2.0 1
3.0 3 3.0 –
2.5 4
2.0 2 2.0 –
2.0 1
3.5 7 1.0 –

| | | | | | |
0 1 2 3 4 5 6 7
Area payroll
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
∑y = 15.0 ∑x = 18 ∑x2 = 80 ∑xy = 51.5

∑xy - nxy 51.5 - (6)(3)(2.5)

x = ∑x/6 = 18/6 = 3 b= = = .25
∑x - nx2
2 80 - (6)(32)

y = ∑y/6 = 15/6 = 2.5 a = y - bx = 2.5 - (.25)(3) = 1.75

 Associative Forecasting Example
^
y = 1.75 + .25x Sales = 1.75 + .25(payroll)
If payroll next year is
estimated to be $6 billion,
then: 4.0 –
Sales = 1.75 + .25(6) 3.25
Sales = $3,250,000 3.0 –

Given our assumption of a DBCC’s sales 2.0 –

Straight line relationship between
Payroll and sales, we now have an 1.0 –
Indication of the slope of that
| | | | | | |
Relationship. Sales increases at 0 1 2 3 4 5 6 7
Rate of a million dollars for every Area payroll
Quarter billion dollars in the local
Payroll. That is because b= .25.
 Standard Error of the Estimate
(A measure of variability around the regression)

 A forecast is just a point estimate of a future

value of y.
 This point is 4.0 –
actually the 3.25
mean of a DBCC’s sales
3.0 –

probability 2.0 –
distribution.
1.0 –

| | | | | | |
0 1 2 3 4 5 6 7
Area payroll
 Correlation Coefficient
 Another way to evaluate relationship of variables,
which measures the degree or strength of the
linear relationship.
 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

nSxy - SxSy
r=
[nSx2 - (Sx)2][nSy2 - (Sy)2]
 Correlation Coefficient
y y

(a) Perfect positive x (b) Positive x

correlation: correlation:
r = +1 0<r<1

y y

(c) No correlation: x (d) Perfect negative x

r=0 correlation:
r = -1
 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
 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
 Monitoring and Controlling
Forecasts

Tracking Cumulative error

signal =
MAD

∑(Actual demand in
period i -
Forecast demand
Tracking in period i)
signal =
(∑|Actual - Forecast|/n)
 Tracking Signal

Signal exceeding limit

Tracking signal
Upper control limit
+

0 MADs Acceptable
range

–
Lower control limit

Time
 Tracking Signal Example
Sipradi Automobiles , Kathmandu wants to evaluate performance
of its Tata Marcopolo forecast, use the forecast and demand data
for the last 6 quarters for Tata Marcopolo sales as given below.
Develop a tracking signal for the forecast and see it stays within
acceptable limit, and the defined limit is ±4 MADs.

Actual Forecast
Qtr Demand Demand

1 90 100
2 95 100
3 115 100
4 100 110
5 125 110
6 140 110
 Tracking Signal Example
Cumulative
Absolute Absolute
Actual Forecast Cumm Forecast Forecast
Qtr Demand Demand Error Error Error Error MAD
1 90 100 -10 -10 10 10 10.0
2 95 100 -5 -15 5 15 7.5
3 115 100 +15 0 15 30 10.0
4 100 110 -10 -10 10 40 10.0
5 125 110 +15 +5 15 55 11.0
6 140 110 +30 +35 30 85 14.2

At the end of quarter 6:

MAD = (∑|Actual - Forecast|/n) = 85/6 = 14.2

Tracking Signal = Cumm Error/ MAD = +35/14.2 = +2.5

Similarly for all the quarters must be calculated and tabulated.

 Tracking Signal Example
Tracking Cumulative
Absolute Absolute
Signal Forecast
Actual Cumm Forecast Forecast
Qtr (Cumm Error/MAD)
Demand Demand Error Error Error Error MAD
1 -10/10
90 = -1 100 -10 -10 10 10 10.0
2 -15/7.5
95 = -2100 -5 -15 5 15 7.5
3 0/10
115 = 0 100 +15 0 15 30 10.0
4 -10/10
100 = -1 110 -10 -10 10 40 10.0
5 +5/11
125 = +0.5110 +15 +5 15 55 11.0
6 +35/14.2
140 = +2.5
110 +30 +35 30 85 14.2

Conclusion ;
The variation of the tracking signal between -2.0 and +2.5 is within
acceptable limits
 A Company’s Rule for Changing the
Smoothing Constant (α )

Limits for Do not Increase α Increase α Increase α

Absolute change by 0.1 by 0.3 by 0.5
Tracking
Signal

0 - 2.4 *
2.5 - 2.9 *
3.0 - 3.9 *
Over 4 *

Note : Usually depends upon subjective opinion of experienced

Decision makers.
 References
• S Buffa – Operations Management
• Jay Heizer, Barry Render – Operations
Management