MADHAV INSTITUTE OF TECHNOLOGY & SCIENCE GWALIOR

120511/Industrial Engineering
Unit-2
Forecasting Quantitative methods
Dr. Nitin Upadhyay
Department of Mechanical Engineering

MITS GWALIOR 1
Source: Source:-https://onlinecourses.nptel.ac.in/noc20_me30/preview

MITS GWALIOR 2
 Forecasting
Methods
Quantitative
• Average, moving average, exponential smoothing

• Causal Methods (Linear regression, multiple regression)

Business Time
Series
 Types of Forecasting
Models
Types of Forecasts
– Qualitative: Based on experience, judgement, knowledge

– Quantitative: Based on data, statistics

- Formal Methods: Systematically reduce forecasting errors;

Time series models

Causal models (e.g. Regression)
 Simple Moving
Average
1
Ft+1 = (Dt + Dt-1 +….+ Dt+1-n )
n
1 t
Ft+1 = å Di
n i=t+1-n

Forecast Ft is average of n previous observations or actuals Dt

 Simple Moving
Average
• Note that the n past observations are equally weighted.
• Issues with moving average forecasts:

– All n past observations treated equally

– Observations older than n are not included at all
– Requires that n past observations be retained
– Problem when 1000's of items are being forecast
 Simple Moving
Average
• Include n most recent observations
• Weight equally
• Ignore older observations Weight
1/n

n ... 3 2 1
Today
Simple Average:
A XYZ television supplier found a demand of 200 sets in
July, 225 sets in August & 245 sets in September. Find the
demand forecast for the month of October using simple
average method.
The average demand for the month of
October is
D1+D2+D3
SA =
3

200+225+245
SA =
3

SA = 223.33
~ 224 units
 Simple Moving Average
A XYZ refrigerator supplier has experienced the following
demand for refrigerator during past five months.

Month Demand
February 20
March 30
April 40
May 60
June 45
 Find out the demand forecast for the month of July
using five-period moving average & three-period
moving average using simple moving average method.
 σ 𝒏𝒊=𝟏𝑫i
MAn =
𝒏
For five period average (i.e. n=5)

𝟐𝟎 + 𝟑𝟎 + 𝟒𝟎 + 𝟔𝟎 + 𝟒𝟓
MA5 =
𝟓
= 39 units
For three period average (i.e. n=3)
MA = 𝟒𝟎 + 𝟔𝟎 + 𝟒𝟓
3
𝟑
= 48.33
≈ 49 units
 Weighted Moving Average Method
The manager of a restaurant wants to make decision on inventory
and overall cost. He wants to forecast demand for some of the items
based on weighted moving average method. For the past three
months he experienced a demand for pizzas as follows:

Month Demand
October 400
November 480
December 550
Find the demand for the month of January by assuming
suitable weights to demand data.

WMA = σ 𝑛𝑖 = 𝐶𝑖 × 𝐷𝑖
Ci = Weights
1 for Periods
Di = Demand for Periods
Let C1 = 0.25, C2 = 0.3, C3 = 0.5
WMA = C1 x D1 + C2 x D2 + C3 x D3
= 0.25 x 400 + 0.3 x 480 + 0.5 x 550
= 100+144+275
= 519 units
 Weighted Moving Average Method :
The past data on the load on the lathe machine is shown
below
Month Demand
May -
June 585
July 610
Aug. 675
Sep. 750
Oct. 860
Nov. 970
Compute a weighted three months moving average for December,
where the weights are 0.5 for the latest month, 0.3 and 0.2 for the
other months respectively.
A three month weighted moving average forecast for December
= (W Nov. x D Nov.) + (W Oct. x D Oct.) + (W Sep. x D Sep.)
= 0.5 x 970 + 0.3 x 860 + 0.2 x 750
= 485+258+150
= 893 machine hours
 Exponential Smoothing
• Include all past observations
• Weigh recent observations much more heavily than very old
observations:
Weight
0 < a <1
Decreasing weight given a
to older observations
a(1-a)
a(1-a)2
a(1-a)3
Today
 Exponential
Smoothing

Ft = aDt + a (1- a )Dt-1 + a (1- a )2 Dt-2 +…

Ft = aDt + (1- a )[aDt-1 + a (1- a)Dt-2 +…]
 Exponential
Smoothing

Ft = aDt + a(1- a)Dt-1 + a(1- a)2 Dt-2 +…

Ft = aDt + (1- a)[aDt-1 + a(1- a)Dt-2 +….]

Or
𝐹% = 𝐹%&' + 𝛼(𝐷%&' − 𝐹%&' )
• Thus, new forecast is weighted sum of old forecast and actual
demand
• Notes:
– Only 2 values (Dt and Ft-1 ) are required, compared with n
for moving average
– Parameter a determined empirically (whatever works best)
– Rule of thumb: a < 0.5
– Typically, a = 0.2 or a = 0.3 work well
• Forecast for k periods into future is:
F t + k = Ft
Exponential Smoothing:
One of the two wheeler manufacturing company experienced
irregular but usually increasing demand for three products. The
demand was found to be 420 bikes for June and 440 bikes for July.

They use a forecasting method which takes average of past year to

forecast future demand. Using the simple average method demand
forecast for June is found as 320 bikes (Use a smoothing coefficient
0.7 to weight the recent demand most heavily) and find the demand
forecast for August.
Exponential Smoothing:
A firm uses simple exponential smoothing with 𝛼 = 0.2 to forecast
demand. The forecast for the first week of January was 400 Units,
whereas actual demand turned out to be 450 Units.
(a) Forecast the demand for the second week of January
(b) Assume that the actual demand during the second week of
January turned out to be 460 units. Forecast the demand upto
February third week, assuming the subsequent demands as 465,
434, 420, 498 and 468 units
 LEAST SQUARE METHOD
(REGRESSION)
• This is the mathematical method of obtaining
"the line of best fit between the dependent
variable (usually demand) and an independent
variable
• In a simple regression analysis, the relationship
between the dependent variable .y and

• some independent variable x can be represented

by a straight line.

𝑦 = 𝑎 + 𝑏𝑥
where b is the slope of the line
a is the y-intercept

MITS GWALIOR 25
The value of a and b are determined by solving simultaneous equations

Σ𝑦 = 𝑛𝑎 + 𝑏Σ𝑥 …(1)
Σ𝑥𝑦 = 𝑎Σ𝑥 + 𝑏Σ𝑥2 ...(2)

Solving equation No. (1). and (2) :

Σ𝑥𝑦 − 𝑛𝑥̅ 𝑦+
𝑏=
Σ𝑥 ! − 𝑛𝑥̅ 2

And 𝑎 = 𝑦+ − 𝑏𝑥̅

MITS GWALIOR 26
 EXAMPLE-1
• The following data gives the sales of the company for various
years. Fit the straight-line. Forecast the sales for the year 1998
and 1999

MITS GWALIOR 27
MITS GWALIOR 28
 Example-2 (Regression)
A company manufacturing washing machines establishes a fact that
there is a relationship between sale of washing machines and
population of the city. The market research carried out reveals the
following information:

Fit a linear regression equation and estimate the demand for

washing machines for a city with a population of 45 million.

MITS GWALIOR 29
MITS GWALIOR 30
MITS GWALIOR 31