2. Definition : A forecast is an estimate of uncertain future event.
Application: it is used to improve decision-making &
planning. Even though forecasts are almost always in error.
Importance in Mechanical engineering: In manufacturing we
need it.
3. They are usually wrong.
Aggregate forecast are usually more accurate.
A good forecast is more than a single number
mean and standard deviation
range (high and low)
Forecasts should not be used to the exclusion of known
information
4. 0-6month
Short term
• Weekly demand
(Inventory control)
• Operational
decision
• Frequently made
• Low level of
responsibility
• Individual items
0.5-2years
Intermediate
• Annual (monthly)
demand (Production
planning)
2+years
Long term
• 10 years Demand
(facility planning)
• Strategic
decision
• Infrequently
made
• Top management
level
• Product line
6. Informed opinion &
judgment
• Subject opinion of
one or more
individuals.
• EXAMPLE: ("grass
roots") collection
and aggregation of
individual sales
forecasts to obtain
overall sales
forecast by product
or region
Delphi method
• An iterative
technique for
obtaining a
consensus forecast
from a group of
experts,
• EXAMPLE: long
range forecasting of
technological
advances
Market Research
• An interview or set of
questions are used to
know the opinion of
potential customers,
current users,&
others.
• EXAMPLE: voter
preference, new car
buyer.
HISTORICAL LIFE-
CYCLE ANALOGY
• Demand for a new
product can be
forecasted by
anticipating an s-
shaped growth curve
similar to the s-curve
experience with
related product.
• EXAMPLE: Product
life cycle curve.
7. 1. Last period
demand
2. Arithmetic
average
Average all
past demand to
average out the
random
fluctuation
3. Simple
Moving
Average (N-
period)
Average of the
N most recent
demands, to
"smooth out"
the random
fluctuations
4. Weighted
Moving
Average (N-
period)
The weighted
moving
average is
equivalent to
the simple
moving
average.
5. Simple
Exponential
Smoothing
A simple way of
calculating a
weighted
moving average
forecast with
exponentially-
declining
weights
6. Calculate
multiplicative
seasonal
indexes
1. Collect monthly
demand data for
several past years
2. For each month
of past data,
calculate the ration
of demand.
Average the ratios
of several years to
get the seasonal
index for the
quarter.
8.
9. The forecast error in period t, et, is the
difference between the forecast for demand in
period t and the actual value of demand in t.
For a multiple step ahead forecast:
et = Ft - t, t - Dt.
For one step ahead forecast: et = Ft -
Dt.
e1, e2, .. , en forecast errors over n
periods
MAD = (1/n) S | ei |
MAPE = [(1/n) S | ei /Di| ]*100
MSE = (1/n) S ei
2
A bias occurs when the average
value of a forecast error tends to
be positive or negative (ie,
deviation from truth).
Mathematically an unbiased
forecast is one in which E (e i ) =
0.
S e i = 0
Biases in Forecasts
Mean Absolute
Deviation
Mean Absolute
Percentage Error
Mean Square
error
10. I have taken the full concept about the forecasting from
the internet specially from
personal.ashland.edu/dlifer/internal/OMlectureforecasting.
pdf & homes.ieu.edu.tr.
I am very much thankful to our faculties of our
department.
I also thankfull to my parents and friends.