Forecasting

“You can never plan the future by the past “…..

Edmund Bruke

“I know of no way of judging the future but by

the past” ……. Patrick Henry
• Your aim is to learn more about
– different types of forecasting
– How are forecasts made
– Why?
– How reliable are they ?
• Key element in operations structure

• Uses data from past events to determine

future events

• Apart from weather fore cast can you think of

three other forecasts?
• Predict world population growth

• Car ownership and demand for roads

• Energy production and consumption

• How is forecast made ????
• Study previous trends
• Identify key variables e.g birth rate, death rate
and migration
• Consider likely changes in these variables over
time e.g. increasing levels of economic
development may lead to higher rates of
energy consumption
• Why is forecasting important ????
• Planning services e.g. schools, houses, jobs
infrastructure, welfare
• Providing resources e.g. food, water, land
energy

• Think about the consequences of a bad

forecast …………… for people, the environment
• Consider world population
– What factors determine the world population ?
• Birth rate
• Death rate
– Population will increase if birth rate is greater than
death rate
• World population growth
– In 1500 the worlds population was 500m
– By 1800 it had slowly grown to 1 billion
– How long had it taken to double
• 300 years
– If you had been a population forecaster in 1800 when
would you have predicted that population would
reach 2 billion ?
• The year2100
• Yet world population was 6 billion by 2000, a six fold
increase
• What went wrong with the
forecast ?????
• Was the model incorrect ?
– No birth rate and death rate were the only
relevant variables
• Were previous trends bad indicator of the
future ?
– Death rate dropped very rapidly over the last 150
years due to better health care, nutrition and
living conditions
• What do the experts forecast?
– World population is currently growing at the rate
1.2% per annum or 77 million people per year
– By 2050 world population is forecast to be
between 7.9 billion ( low variant) and10.9 billion
(high variant), and a medium variant of 9.3 billion
– Some experts forecast that world population
could reach 12 – 15 billion before leveling off
• Why do these forecast vary so
much ????
• Regional growth rates differ
– Africa is supposed to double its population by
2050
• What may changer this forecast ?
– Higher death rates e.g. AIDS
– China and India currently compromise 30% of the
worlds population
• Which country is forecast to grow at the fastest rate by
2050
– Both are under going rapid development and industrialization
…………….
• So to take another example
– How might the above effect fuel consumptions
– What are the variables … too many to list but
such things as
• Availability of fuel resources
• Demand which does not depend only on size of
population also economic development, geographical
location and more
• Capital and technology
• Politics, trade , international agreements ………. More!!!
• What can be forecast with
certainty ????
• Tides ?
• Solar and lunar eclipses ?
• Transits of Venus ?
• Never the less
– Forecasting is critical to improving the competitive
edge
– It enables a company to respond quickly and
accurately to market change than would
otherwise be possible
• How does forecasting relate to
the management process of
planning , organizing and
controlling?
• If operations have been properly planned and
organized control is easier and smoother …..
These processes interrelate and overlap …..
• Forecasting reduces costs of adjusting
operations in response to unexpected
deviations by specifying future demand.
• Accurate forecasts increase operating
efficiency
• Managers may want long run estimates of
over all demand or short run estimates of
demand for each individual products
• A forecast is an estimate of a future event
achieved by systematically combining and
casting forward in a predetermined way data
about the past
• A prediction is an estimate of a future event
achieved thru subjective consideration other
than just past data. This subjective
consideration need not occur in any
predetermined way.
• For predicting good subjective estimates can
be based on managers skill, experience and
judgment
• Forecasting requires statistical and
management science techniques
• Forecasts are necessary for planning,
scheduling and controlling the system to
facilitate effective and efficient output of
goods and services
 Information on most recent demand
and production

Demand forecast
for operations

Planning the system

(designing) Scheduling the system Controlling the
Product Aggregate production system
Process equipment
planning Production
investment
Replacement Operation scheduling inventory labor
Capacity planning
cost

Output of goods
and services
• Characteristics of demand over
time
• To systematically analyze data for forecasting a time series
analysis is used it could be
– Constant
– Seasonal
– linear trend
– Or combination of the above
• Noise – term used to define the data points that fall on the
pattern (even if not exactly they cluster around the pattern)
• High noise many of the points lie far away from the pattern
and low is the reverse
 Seasonal
(cyclical) Linear trend

production
demand

constant

time