Operations Management

Lect. 7 “Forecasting”

Dr: Aly Hassan Abdelbaky Elbatran

Head of Mechanical Engineering (S_V)

Faculty of Engineering, Arab Academy for Science
and Technology and Maritime Transport, Egypt

1
 Forecasting Principles

Forecasting:- Predicting future events.

1. Forecasts are rarely perfect:- Forecasting the future involves

uncertainty. Therefore, it is almost impossible to make a

perfect prediction.

2. Forecasts are more accurate for grouped data than for

individual items

3. Forecast are more accurate for shorter than longer time

periods

2
STEPS IN THE FORECASTING PROCESS

1. Decide what needs to be forecast

Level of detail, units of analysis & time horizon required

2. Evaluate and analyze appropriate data

Identify needed data & whether it’s available

3. Select and test the forecasting model

Cost, ease of use & accuracy

4. Generate the forecast

5. Monitor forecast accuracy over time

3
 Types of Forecasting Methods

➢ Qualitative methods: – judgmental methods

➢ Forecasts generated subjectively by the forecaster

➢ Educated guesses

➢ Quantitative methods: – based on mathematical

modeling:

➢ Forecasts generated through mathematical

modeling

4
Types of Forecasting Methods

5
 Qualitative Method

Type Characteristics Strengths Weaknesses

Executive A group of Good for strategic One person's
opinion managers meet & or new-product opinion can
come up with a forecasting dominate the
forecast forecast
Market research Uses surveys & Good determinant It can be difficult to
interviews to of customer develop a good
identify customer preferences questionnaire
preferences
Delphi method Seeks to develop Excellent for Time consuming to
a consensus forecasting long- develop
among a group of term product
experts demand,
technological
changes, and
scientific advances

6
 Quantitative Method

➢ Time Series Models:

➢ Assumes information needed to generate a forecast is

contained in a time series of data

➢ Assumes the future will follow same patterns as the past

➢ Causal Models or Associative Models

➢ Explores cause-and-effect relationships

➢ Uses leading indicators to predict the future

➢ Housing starts and appliance sales

7
 Time Series Method

➢ Forecaster looks for data patterns as

➢ Data = historic pattern + random variation

➢ Historic pattern to be forecasted:

➢ Level (long-term average) – data fluctuates around a constant

mean

➢ Trend – data exhibits an increasing or decreasing pattern

➢ Seasonality – any pattern that regularly repeats itself and is of a

constant length

➢ Cycle – patterns created by economic fluctuations

➢ Random Variation cannot be predicted

8
Time Series Pattern

9
 Time Series Models
➢ Naive:

➢ The forecast is equal to the actual value observed during the last period –
good for level patterns

Ft +1 = At
where Ft+1 = forecast for next period, t + 1

At = actual value for current period, t

t = current time period

o For example, suppose that we have monthly seasonality and know that sales for last
January were 230 units. Using the naïve method, we would forecast sales of 230 units
for next January.

10
 Time Series Models
➢ Simple Mean:

➢ The average of all available data - good for level patterns

Ft +1 =  A t / n

where Ft+1 = forecast of demand for next period, t + 1

At = actual value for current period, t
n = number of periods or data points to be averaged

11
 Time Series Models
Moving Average:

➢ The average value over a set time period (e.g.: the last four weeks)

➢ Each new forecast drops the oldest data point & adds a new observation

➢ More responsive to a trend but still lags behind actual data

Ft +1 =  A t / n

50+52+48
= 50
3

12
 Time Series Models

Three-point Moving Average

Five-point Moving Average

13
Time Series Models

14
 Time Series Models
Weighted Moving Average:

➢ All weights must add to 100% or 1.00 (e.g. Ct 0.5, Ct-1 0.3, Ct-2
0.2 (weights add to 1.0)

➢ Allows emphasizing one period over others; above indicates more

weight on recent data (Ct=0.5)

➢ Differs from the simple moving average that weighs all periods
equally - more responsive to trends

15
Time Series Models

16
 Time Series Models
Exponential Smoothing:

➢ Most frequently used time series method because of ease of use and minimal
amount of data needed

➢ Need just three pieces of data to start:

➢ Last period’s forecast (Ft)

➢ Last periods actual value (At)

➢ Select value of smoothing coefficient, α ,between 0 and 1.0

➢ If no last period forecast is available, average the last few periods or use naive
method

➢ Higher α values (e.g. .7 or .8) may place too much weight on last period’s random
variation

Ft +1 = αA t + (1 − α )Ft

17
 Time Series Models
The Hot Tamale Mexican restaurant uses exponential smoothing to forecast monthly usage of
tabasco sauce. Its forecast for September was 200 bottles, whereas actual usage in September was
300 bottles. If the restaurant’s managers use an of 0.70, what is their forecast for October?

18
 Time Series Models

Determine forecast for periods 7 & 8

Period Actual
- 2-period moving average 1 300
- 4-period moving average 2 315
3 290
- 2-period weighted moving average with t-
4 345
1 weighted 0.6 and t-2 weighted 0.4 5 320
6 360
- Exponential smoothing with alpha=0.2
7 375
and the period 6 forecast being 375 8

19
 Time Series Models

Period Actual 2-Period 4-Period 2-Per.Weighted. Expon. Smooth.

1 300

2 315

3 290

4 345

5 320

6 360

7 375 340.0 328.75 344.0 372.0

8 367.5 350.0 369.0 372.6

20
 Forecasting Trend
Basic forecasting models for trends compensate for the lagging that would
otherwise occur
Trend-adjusted exponential smoothing:- Exponential smoothing model that is
suited to data that exhibit a trend.

21
 Forecasting Trend
Forecasting trend problem: a company uses exponential smoothing with trend to forecast
usage of its lawn care products. At the end of July the company wishes to forecast sales
for August. July demand was 62. The trend through June has been 15 additional gallons of
product sold per month. Average sales have been 57 gallons per month. The company
uses alpha+0.2 and beta +0.10. Forecast for August.
➢ Smooth the level of the series:

S July = αA t + (1 − α)(S t −1 + Tt −1 ) = (0.2)(62) + (0.8)(57 + 15) = 70

➢ Smooth the trend:

TJuly = β(S t − St −1 ) + (1 − β)Tt −1 = (0.1)(70 − 57 ) + (0.9)(15) = 14.8

➢ Forecast including trend:

FITAugust = St + Tt = 70 + 14.8 = 84.8 gallons

22
 Liner Trend line
A time series technique that computes a forecast with trend by drawing a

straight line through a set of data using this formula:

Y = a + bX

where

Y = forecast for period X

X = the number of time periods from X = 0

a = value of y at X = 0 (Y intercept)

b = slope of the line

23
 Liner Trend line
A manufacturer has plotted product sales over the past four weeks. Use a linear trend line to
generate forecast for week 5.
Week Sales
1 2300
2 2400
2375
3 2300

b=
 XY − n XY =24,000−4(2.5)(2375) = 50 4 2500
30−4(2.52 )
 X − nX
2 2

Week (X) Sales (Y) 𝑿𝟐 XY

1 2300 1 2300
2 2400 4 4800
3 2300 9 6900
4 2500 16 10000
10 9500 30 24000
24
 Forecasting Seasonality

1. Calculate the average demand per season

E.g.: average quarterly demand

2. Calculate a seasonal index for each season of each year:

Divide the actual demand of each season by the average

demand per season for that year

3. Average the indexes by season

E.g.: take the average of all Spring indexes, then of all Summer

indexes, ...

25
 Forecasting Seasonality

4. Forecast demand for the next year & divide by the number of

seasons

➢ Use regular forecasting method & divide by four for average

quarterly demand

5. Multiply next year’s average seasonal demand by each average

seasonal index

➢ Result is a forecast of demand for each season of next year

26
 Forecasting Seasonality
University wants to develop forecasts for next year’s quarterly enrollment. It has collected quarterly
enrollments for the past two years. It has also forecast total annual enrollment for next year to be
90,000 students. What is the forecast for each quarter of next year?

Quarter Year 1 Year 2

Fall 24,000 26,000

Winter 23,000 22,000

Spring 19,000 19,000

Summer 14,000 17,000

Total 80,000 84,000

27
 Forecasting Seasonality
Step 2 Step 3 Step 4 Step 5

Quarter Year 1 Seasonal Year 2 Seasonal Avg. Year3

Index Index Index
Fall 24000 1.2 26000 1.238 1.219 27428

Winter 23000 1.15 22000 1.048 1.099 24728

Spring 19000 0.95 19000 0.905 0.928 20,880

Summer 14000 0.7 17000 0.810 0.755 16,988

Total 80000 84000 90000

Average 20000 21000 22500

Demand
Step 1

28
 Casual Models
◼ Often, leading indicators can help to predict changes in future

demand e.g. housing starts

◼ Causal models establish a cause-and-effect relationship between

independent and dependent variables

◼ A common tool of causal modeling is linear regression: Y= a + bX

◼ Additional related variables may require multiple regression

modeling

29
 Casual Models

◼ Identify dependent (y) and

independent (x) variables

◼ Solve for the slope of the line

b=
 XY − n X Y

 X − nX
2 2

◼ Solve for the y intercept

a = Y − bX
◼ Develop your equation for
the trend line

Y = a + bX

30
 Casual Models
A maker of personalized golf shirts has been tracking the relationship between
sales and advertising dollars over the past four years. The results are as
follows:
Sales ($) Advertising ($)

130,000 32,000

151,000 52,000

150,000 50,000

158,000 55,000

Use linear regression to find out what sales would be if the company invested
$53,000 in advertising for next year.

31
 Casual Models
Sales in Adv. in XY 𝑋2 𝑌2
Thousan Thous
b=
 XY − n XY
ds $ (Y) ands $
 X − nX 2 2

(X)

1 130 32 4160 2304 16,900

2 151 52 7852 2704 22,801 28202 − 4(47.25 )(147.25 )

b= = 1.15
9253 − 4(47.25 )
2
3 150 50 7500 2500 22,500
a = Y − b X = 147.25 − 1.15(47.25 )
4 158 55 8690 3025 24964 a = 92.9
5 153.85 53 Y = a + bX = 92.9 + 1.15X
Y = 92.9 + 1.15(53 ) = 153.85
Tot 589 189 28202 9253 87165

Avg 147.25 47.25

32
 Correlation Coefficient
➢ Correlation coefficient (r) measures the direction and strength of the linear
relationship between two variables. The closer the r value is to 1.0 the better
the regression line fits the data points.

n ( XY ) − ( X )( Y )
r=
( X ) − ( X ) ( Y ) − (Y )
2 2
2 2
n * n
4(28,202 ) − 189(589 )
r= = .982
4(9253) - (189) * 4(87,165) − (589 )
2 2

r 2 = (.982 ) = .964
2

➢ Coefficient of determination ( 𝑟 2 ) measures the amount of variation in the

dependent variable about its mean that is explained by the regression line.
Values of ( 𝑟 2 ) close to 1.0 are desirable.
33
 Casual Models

➢ An extension of linear regression but:

➢ Multiple regression develops a relationship between a

dependent variable and multiple independent variables. The
general formula is:

34
 Forecast Accuracy Measure

➢ Forecasts are never perfect

➢ Need to know how much we should rely on our chosen

forecasting method

➢ Measuring forecast error:

E t = A t − Ft
where
Et forecast error for period t
At actual value for period t
Ft forecast for period t

➢ Note that over-forecasts = negative errors and under-

forecasts = positive errors

35
 Forecast Accuracy Measure

➢ Mean Absolute Deviation (MAD)

➢ measures the total error in a forecast without regard to sign

MAD =
 actual − forecast
n
➢ Cumulative Forecast Error (CFE)
➢ Measures any bias in the forecast

CFE =  (actual − forecast )

➢ Mean Square Error (MSE)

➢ Penalizes larger errors
 (actual - forecast )2

MSE =
n
➢ Tracking Signal
➢ Measures if your model is working

CFE
TS =
M AD
36
 Forecast Accuracy Measure
Standard Parts Corporation is comparing the accuracy of two methods
that it has used to forecast sales of its popular valve. Forecasts using
method A and method B are shown against the actual values for January
through May. Which method provided better forecast accuracy?
Method A Method B
Month Actual F’cast Error Absolute 𝐸𝑟𝑟𝑜𝑟 2 F’cast Error Absolute 𝐸𝑟𝑟𝑜𝑟 2
sales Error Error

Jan. 30 28 2 2 4 27 3 3 9

Feb. 26 25 1 1 1 25 1 1 1
March 32 32 0 0 0 29 3 3 9
April 29 30 -1 1 1 27 2 2 4
May 31 30 1 1 1 29 2 2 4
Total 3 5 7 11 11 27
MAD 5/5=1 11/5=2.2

MSE 7/5=1.4 27/5=5.4

37
 Forecast Accuracy Measure
A company uses a tracking signal with limits of ± 4 to decide whether a
forecast should be reviewed. Compute the tracking signal given the following
historical information and decide when the forecast should be reviewed.

Method A Method B
Month Actual F’cast Error Accumul Tracking F’cast Error Accumul Tracking
sales ative Signal ative Signal
Error Error

Jan. 30 28 2 2 2 27 3 3 1.5
Feb. 26 25 1 3 3 25 1 4 2
March 32 32 0 3 3 29 3 7 3.5
April 29 30 -1 2 2 27 2 9 4.5
May 31 30 1 3 3 29 2 11 5.5
MAD 5/5=1 10/5=2
MSE 7/5=1.4 27/5=5.4

38
 Selecting the Right Forecasting Model
▪ The amount & type of available data

▪ Some methods require more data than others

▪ Degree of accuracy required

▪ Increasing accuracy means more data

▪ Length of forecast horizon

▪ Different models for 3 month vs. 10 years

▪ Presence of data patterns

▪ Lagging will occur when a forecasting model meant for a

level pattern is applied with a trend

39
THANK YOU

40