Chapter 3

FORECASTING
(metrology, population, price, growth, demand forecast
etc….)
 What is forecasting?
 Predictions, projections or estimates of future events

 usually as a result of rational study or analysis of relevant data.

 Objective of forecasting: predicting future by reducing the forecast error.

The purpose of forecasting is to guide organizational goals such as:
 Production
 Inventory
 Personnel
 Facilities
 Forecasting time horizons
Short-range forecast
 generally less than 3 months
 Purchasing, job scheduling, workforce levels, job assignments, production
levels
 Short-term forecasts tend to be more accurate than longer-term forecasts
Medium-range forecast
 3 months to 3 years
 Sales and production planning, budgeting
Long-range forecast
 3+ years
 New product planning, facility location, research and development(R&D)
 Medium/long range forecasts deal with more comprehensive issues and support
management decisions regarding planning and products, plants and processes
 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
 Product life cycle (the life of a product in the market with
respect to business/commercial costs and sales measures)

Introduction Growth Maturity Decline

 Product design and  Forecasting critical  Little product

development critical  Standardization differentiation
OM Strategy/Issues

 Product and process

 Frequent product reliability  Optimum capacity  Cost
process design  Competitive product minimization
 Increasing stability
changes improvements of process  reduce line to
 High production  Increase capacity  Long production
eliminate items
costs not returning
runs good margin
 Attention to quality
 Reduce capacity
 Product
improvement and
cost cutting
 Types of forecasts.

Economic forecasts

 Address business cycle – inflation rate, money supply, etc.

Technological forecasts

 Predict rate of technological progress

 Impacts development of new products

Demand forecasts

 Predict sales of existing products and services

 The realities!

 Forecasts are rarely perfect(can be higher or lower than the realization).

 Product family and aggregated forecasts are more accurate than individual product
forecasts
 We usually prefer simple models than complex ones in forecasting

Reasons for preferring simple models:

 Easier to estimate precisely
 Easier to interpret, understand
 Easier to communicate
 Forecasting approaches

1. Qualitative Methods

 Used when situation is unclear and little data exist about

• New products
• New technology
 spontaneous, largely educated guesses
 Usually cannot be reproduced by someone else
 Not a precise approach
 May be quite appropriate in certain situations
Classifications of qualitative methods

1. Jury of executive opinion

2. Delphi method

3. Sales force composite

4. Consumer Market Survey

 1. Jury of executive opinion
 Involves small group of high-level experts
 Group estimates demand by working together
 Combines managerial experience with statistical models
 Relatively quick

2. Delphi method
 Iterative group process, continues until
consensus is reached
 3 types of participants
 Staff
 Respondents
 Decision makers
 3. Sales force composite

 Each salesperson projects his or her sales

 Combined at district and national levels
 Tends to be overly optimistic

4. Consumer market survey

 Ask customers about purchasing plans.

 What consumers say, and what they actually do are

often different.

 Sometimes difficult to answer.

Forecasting approaches

2. Quantitative Methods

 Used when situation is ‘stable’ and historical data exist about

• Existing products
• Current technology
 Based on mathematical or statistical models
 Have underlying model or technique
 Results produced automatically from the model
 Fully reproducible by any forecaster
Overview of quantitative approaches

1. Naive approach
2. Moving averages Time-series models
3. Exponential smoothing
4. Trend projection
5. Linear regression Associative
model
 Time series forecasting
Set of evenly spaced numerical data
* Obtained by observing response variable at regular time periods
Forecast based only on past values, no other variables important
* Assumes that factors influencing past and present will continue influence in future

Time Series
The forecast statement of time series can be:
 Point forecast: a single number
 Interval forecast: an interval
 Density forecast: the probability density
1. 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
 Sometimes cost effective and efficient
 Can be good starting point
2a. Moving average method
MA is a series of arithmetic means
In words: the arithmetic average of the n most recent observations. For a one-step-ahead
forecast:
Ft = (1/N) (Dt - 1 + Dt - 2 + . . . + Dt - n )

∑ demand in previous n periods

Moving average = n
Moving Average Example

Actual 3-Month
Month Shed Sales Moving Average

January 10
February 12
March 13
April 16 (10 + 12 + 13)/3 = 11.67
May 19 (12 + 13 + 16)/3 = 13.67
June 23 (13 + 16 + 19)/3 = 16
July 26 (16 + 19 + 23)/3 = 19.33
 Moving Average - example
MONTH Demand Month Demand
3 month MA:
January 89 July 223 =(oct+nov+dec)/3=258.33
February 57 August 286

March 144 September 212 6 month MA:

April 221 October 275

=(jul+aug+…+dec)/6=249.33

May 177 November 188

12 month MA:
June 280 December 312
=(Jan+feb+…+dec)/12=205.33

4 - 17
Graph of Moving Average

30 –
28 – Moving
26 – Average
– Forecast
24 Actual
22 – Sales
Sales

20 –
18 –
16 –
14 –
12 –
10 –
| | | | | | | | | | | |
J F M A M J J A S O N D
2b. Weighted moving average
• Used when some trend might be present
• Older data usually less important
• Weights based on experience and intuition
Weighted ∑ (Weight for period n) x
Moving Average = (Demand in period n)
∑ Weights
This method looks at past data and tries to logically attach importance to certain data over other
data. Can weight recent higher than older. For example:
If forecasting staffing, we could use data from the last four weeks where Tuesdays are to be
forecast.
Weighting on Tuesdays is: T-1 is .25; T-2 is .20; T-3 is .15; T-4 is .10 and Average of all other days
is weighed .30.
WEIGHTED MOVING AVERAGE Weights Applied Period
3 Last month
2 Two months ago
1 Three months ago
6 Sum of weights

Actual 3-Month Weighted

Month Shed 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
Potential problems with moving average
 Increasing n smoothes the forecast but makes it less sensitive to changes.
 Do not forecast trends well.
 Require extensive historical data.
Moving average and weighted moving average
 3. Exponential smoothing
 Form of weighted moving average
• Weights decline exponentially
• Most recent data weighted most
 Requires smoothing constant (α)
• Ranges from 0 to 1
• Subjectively chosen
 Involves little record keeping of past data
 Exponential smoothing

New forecast = Last period’s forecast + α (Last period’s actual demand– Last period’s forecast)

Ft = Ft – 1 + α(At – 1 - Ft – 1)

Where:- Ft = new forecast

A = previous actual demand
Ft – 1 = previous forecast
α = smoothing (or weighting) constant
(0 ≤ a ≤ 1)
 Exponential smoothing example

 Predicted demand = 142 Ford

 Actual demand = 153
 Smoothing constant a = 0.20
New forecast = 142 + 0.2(153 – 142)
= 142 + 2.2
= 144.2 ≈ 144 cars
Ft = Ft – 1 + α(At – 1 - Ft – 1)

New forecast = Last period’s forecast + α (Last period’s actual demand– Last period’s forecast)
 Impact of ’s

225 –

200 – Actual
demand a = 0.5
Demand

175 –

150 – a = 0.1
| | | | | | | | |
1 2 3 4 5 6 7 8 9
Quarter
 Impact of ’s

225 –

200 –
 Chose high values of  when underlying
Actual
average isdemand
likely to change  = 0.5
Demand

 Choose low values of  when underlying

175 –
average is stable

150 –  = 0.1
| | | | | | | | |
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

Mean Absolute Deviation (MAD)

MAD = ∑ |Actual - Forecast|
n

Mean Squared Error (MSE)

∑ (Forecast Errors)2
MSE =
n
COMMON MEASURES OF ERROR

 Mean Absolute Percent Error (MAPE)

n
∑[100|Actuali - Forecasti|]/Actuali
MAPE = i = 1

n
NEXT CLASS

4 - 30
 COMPARISON OF FORECAST ERROR Ft = Ft – 1 + α(At – 1 - Ft – 1)
Rounded Absolute Rounded Absolute
Actual Forecast Deviation Forecast Deviation
Tonnage with for with for
Quarter Unloaded  = 0.10  = 0.10  = 0.50  = 0.50
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
175+(180-175)x0.1=175.5 82.45 98.62
COMPARISON OF FORECAST ERROR

Rounded Absolute Rounded Absolute

Actual Forecast Deviation Forecast Deviation
Tonnage∑ |deviations|
with for with for
MAD
Quarter =
Unloaded a = .10 a = .10  = 0.50  = 0.50
n
1 For = 0.10175
180 5.00 175 5.00
2 = 82.45/8
168 175.5 = 10.31
7.50 177.50 9.50
3 For 159
= 0.50174.75 15.75 172.75 13.75
4 175 173.18 1.82 165.88 9.12
5
=
190
98.62/8
173.36
= 12.33
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
COMPARISON OF FORECAST ERROR
Rounded Absolute Rounded Absolute
∑ (forecast
Actual Forecast errors) 2
Deviation Forecast Deviation
MSE =Tonnage with for with for
Quarter Unloaded n
a = .10 a = .10  = 0.50  = 0.50
1 180 175 5.00 175 5.00
For
2  =168
0.10 175.5 7.50 177.50 9.50
3 =159
1,526.54/8
174.75 = 190.82
15.75 172.75 13.75
For
4  =175
0.50 173.18 1.82 165.88 9.12
5
6
=190
205
173.36
1,561.91/8
175.02
= 16.64
195.24
29.98
170.44
180.22
19.56
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
Mean Absolute Deviation (MAD)
MAD 10.31 12.33
 COMPARISON OF FORECAST ERROR
Rounded Absolute Rounded Absolute
Actual Forecast Deviation Forecast Deviation
Tonnage with for with for
Quarter Unloaded  = 0.10  = 0.10  = 0.50  = 0.50
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
(180-175)x0.1+175=175.5 82.45 98.62
4 - 34
COMPARISON OF FORECAST ERROR mean absolute percent error

n Rounded Absolute Rounded Absolute

Actual
Tonnage ∑
Forecast
with
Deviation
100|deviation i
for
Forecast
|/actual i
with
Deviation
for
MAPE Unloaded
Quarter = i = 1 a = .10 a = .10 a = .50  = 0.50
1 180 175 n 5.00 175 5.00
2 For 168
 = 0.10
175.5 7.50 177.50 9.50
3 159 174.75
4 175
= 44.75/8
173.18
= 15.75
5.59%
1.82
172.75
165.88
13.75
9.12
5 For 190
 = 0.50
173.36 16.64 170.44 19.56
6 205 = 54.05/8
175.02 = 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
Mean Absolute Deviation (MAD) MAD 10.31 12.33
Mean Squared Error MSE 190.82 195.24
 mean absolute percent error

[(∑ | Actual demand − forecast |

---------------------------------------- × 100
Actual demand)
MAPE = ---------------------------------------------------------------------
N.

4 - 36
COMPARISON OF FORECAST ERROR
Actual Rounded Absolute Rounded Absolute
Tonnage Forecast Deviation Forecast Deviation
Quarter Unloaded  = 0.10  = 0.10  = 0.50  = 0.50
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
Mean Absolute Deviation ) MAD 10.31 12.33
Mean Squared Error MSE 190.82 195.24
 Mean Absolute Percent Error MAPE 5.59% 6.76%