Industrial Management and Engineering Economics

Kinfe F. Weldetnsae
Samara University,
College of Engineering and Technology,
Department of Mechanical Engineering

April 2022
 OUTLINE

1 Forecasting
Introduction to Forecasting
Forecasting Range
Forecasting Techniques

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 1 / 44
 Introduction to Forecasting
A statement about the future value of a variable of interest such as demand and a
tool used for predicting future value (demand) based on past information.
Process of predicting a future event underlying basis of all business decisions:
Production
Inventory
Personnel
Facilities

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 2 / 44
 Importance of Forecasting
Marketing managers:
Use sales forecasts to determine optimal sales force allocations.
Set sales goals.
Plan promotions and advertising.
Planning for capital investments:
Predictions about future economic activity.
Estimating cash inflows accruing from the investment.
The personnel department:
Planning for human resources.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 3 / 44
 Importance of Forecasting
Managers of nonprofit institutions:
Forecasts for budgeting purposes.
Universities:
Forecast student enrollments.
Cost of operations.
The bank has to forecast:
Demands of various loans and deposits
Money and credit conditions so that it can determine the cost of money it lends.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 4 / 44
 Importance of Forecasting
Manufacturers:
Worker absenteeism
Machine availability
Material costs
Transportation and production lead times, etc.
Service providers:
Forecasts of population
Demographic variables
Weather, etc.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 5 / 44
 Principles of Forecasting
Many types of forecasting models that differ in complexity and amount of data and
way they generate forecasts:
Forecasts rarely perfect because of randomness.
Forecasts more accurate for groups vs. individuals.
Forecast accuracy decreases as time horizon increases.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 6 / 44
 Forecasting Range
Short-range forecasts: Usually less than 3 months.
Concerned with the daily operations of a business firm.
Daily demand or resource requirements, Job scheduling and worker assignment.
A medium-range forecast:
Encompasses anywhere from 3 months to 2 years.
Yearly production and Sales planning and reflect peaks and valleys in demand and
the necessity to secure additional resources for the upcoming year.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 7 / 44
 Forecasting Range
A long-range forecast:
Encompasses a period longer than 1 or 2 years.
Long-range forecasts are related to management’s attempt to:
Plan new products for changing markets.
Build new facilities.
Secure long-term financing

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 8 / 44
 Steps of Forecasting
1 Decide what needs to be forecasted.
Level of detail, units of analysis and time horizon required.
2 Evaluate and analyze appropriate data.
Identify needed data and whether it’s available.
3 Select and test the forecasting model.
Cost, ease of use and accuracy.
4 Generate the forecast.
5 Monitor forecast accuracy over time.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 9 / 44
 Forecasting Techniques
Types of forecasts:
Economic forecasts
Address business cycle – inflation rate, money supply, and etc.
Technological forecasts
Predict rate of technological progress.
Impacts development of new products.
Demand forecasts
Predict sales of existing product.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 10 / 44
 Forecasting Techniques
Qualitative Methods
Used when situation is vague and little data exist.
New products
New technology
Innovative products
Involves intuition, experience.
Forecasts generated subjectively by the forecaster.
Educated guesses.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 11 / 44
 Forecasting Techniques

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 12 / 44
 Forecasting Techniques
Qualitative Forecasting Methods
Individual Expert:
Individual market experts can be hired to watch for industry trends, to estimate
future demand for products.
Executive Opinions/Group Consensus:
The subjective views of executives or experts from sales, production, finance,
purchasing, and administration are averaged to generate a forecast about future
sales.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 13 / 44
 Forecasting Techniques
Qualitative Forecasting Methods
Delphi Method:
Based on sequential questionnaires.
Requires one person to administer and coordinate the process and poll the team
members (respondents) through a series of sequential questionnaires.
Consumer Surveys:
Surveys regarding specific consumer purchases.
Surveys may consist of telephone contacts, personal interviews, or questionnaires as
a means of obtaining data.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 14 / 44
 Forecasting Techniques
Qualitative Forecasting Methods
Quantitative analysis typically involves two approaches:
Causal models
Time-series methods
Causal/Regression Methods:
Causal models establish a quantitative link between observable or known variable
(like advertising expenditures) with the demand for some product.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 15 / 44
 Forecasting Techniques
Causal Models
Causal models establish a cause-and-effect relation ship between independent and
dependent variables.
A common tool of causal modeling is linear regression:
Additional related variables may require multiple regression modeling.
Y=a+bX (1)

Where:
Y- Dependent Variable
X- Independent Variable

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 16 / 44
 Forecasting Techniques
Identify dependent (Y) and
independent (X) variables
Solve for the slope of the line:
Pn
XY − nX̄ Ȳ
b = P1n 2 2
(2)
1 X − n X̄

Solve for the y intercept

a = Ȳ − b X̄ (3)

Develop your equation for the trend

line:
Y = a + bX (4)
Kinfe F. Weldetnsae (Samara University, College of Engineering
Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 17 / 44
 Linear Regression- Example
A company has been tracking the relationship between sales and advertising dollars. Use
linear regression to find out what sales might be if the company invested $53,000in
advertising next year using the following previous data.

Solution
Pn
XY − nX̄ Ȳ
b= P1n 2 2
1 X − n X̄
28202 − 4(47.25)(147.25)
b= =1.15
9253 − 4(47.25)2
a=Ȳ +bX̄ = 147.25 -1.15( 47.25)=92.9
Y=a+bX=92.9+1.15(47.25)=153.85

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 18 / 44
 Correlation Coefficient
Correlation coefficient (r) measures the direction and strength of the linear relationship
between two variables. The closer their value is to 1.0 the better the regression line fits
the data points.

n( n1 XY ) − ( n1 X )( n1 Y )
P P P
r=q P (5)
n 2
Pn 2 p Pn 2 Pn 2
n( 1 X ) − ( 1 X ) n( 1 Y ) − ( 1 Y )

4(28, 202) − 189(589))

r= p p =.982
4(9253 − 1892 ) 4(87, 165) − (589)2
r 2 = .9822 =.964

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 19 / 44
 Time Series Forecasting Methods
Time series forecasting methods are:
Based on analysis of historical data.
Set of evenly spaced numerical data:
Obtained by observing response variable at regular time periods.
Forecast based only on past values:
Assumes that factors influencing past and present will continue influence in future.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 20 / 44
 Time Series Forecasting Methods

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 21 / 44
 Time Series Forecasting Methods
Historic data may exhibit one of the following pattern:
Level (long-term average) – data fluctuates around a constant mean.
Trend – data exhibits an increasing or decreasing pattern.
Seasonality – effects are similar variations occurring during corresponding periods,
can be quarterly, monthly,weekly, daily, or even hourly indexes.
Cycle – are the long-term swings about the trend line.
Irregular variations - caused by unusual circumstances.
Random variations - caused by chance, cannot be predicted.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 22 / 44
 Time Series Forecasting Methods

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 23 / 44
 Time Series Models
Time Series : a set of observations measured at successive times or over successive
periods.
Naı̈ve or Projection
Simple Moving Average
Weighted Moving Average
Exponential Smoothing

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 24 / 44
 Time Series Models
Naı̈ve or Projection
The forecast for the period t, Ft, is simply a projection of previous period t-1
demand, At−1 .
Ft = At−1
E.g. If the actual demand of period t is 120, then the forecast of the period t+1 is
120.
This method, although easy to use, doesn’t make use of datathat is easily available
to most managers; thus, using more ofthe historical data should improve the
forecast.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 25 / 44
 Time Series Models
Simple Moving Average (MA)
An n-period moving average uses the last n periods of demand as a forecast for
next periods demand:.
Ft = At−1 + At−2 + At−3 + ...... + At−n
Where n = total number of periods in the average
Ft =Forecast for period t,
At−1 ,At−2 ,At−3 ,....,At−n =Actual value for periods 1, 2,. . . , n.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 26 / 44
 Time Series Models
Simple Moving Average (MA)
To determine the length of n:
Higher value of n - greater smoothing, lower responsiveness. item Lower value of n -
less smoothing, more responsiveness. item A large value of n is appropriate if the
underlying pattern of demand is stable.
A smaller value of n is appropriate if the underlying pattern is changing or if it is
important to identify short-term fluctuations.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 27 / 44
 Time Series Models
Simple Moving Average (MA)
Example: A company sells storage shed, Determine the forecast of January using
3 month simple moving average.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 28 / 44
 Time Series Models
Weighted Moving Average:
A weighted moving average is a moving average where each historical demand may
be weighted differently.
This runs counter to ones intuition that the most recent data is the most relevant.
Thus, the weighted moving average allows for more emphasis to be placed on the
most recent data. This forecast is:
Wt−1 At−1 + Wt−2 At−2 + Wt−3 At−3 + ...... + Wt−n At−n
Ft = (6)
Wt−1 + Wt−2 Wt−3 + .... + Wt−n

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 29 / 44
 Time Series Models
Weighted Moving Average:
Where wt−1 is the weight applied to the actual demand incurred during period t-1,
and so on.
Intuitively, the expectation would be that the more recent demand data should be
weighted more heavily than older data; so, generally, one would expect the weights
to follow the relationship Wt ≥ Wt−1 ≥ Wt−2 ≥ . . . .Wt−n
The sum of the weights is 1.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 30 / 44
 Time Series Models
Weighted Moving Average:
Example: Consider the weights 3/6, 2/6, 1/6for periods t-1, t-2 and t-3
respectively which are added to one.Determine the forecast of January.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 31 / 44
 Time Series Models
Exponential smoothing:
Nice properties of a weighted moving average would be one where the weights not
only decrease as older and older data are used, but one where the differences
between the weights are “smooth”.
Obviously the desire would be for the weight on the most recent data to be the
largest.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 32 / 44
 Time Series Models
Exponential smoothing:
The weights should then get progressively smaller the more periods one considers
into the past.
The exponentially decreasing weights of exponential smoothing forecast fit this bill
nicely.
Ft = αAt−1 + (1 − α)Ft − 1 (7)
Ft = Ft−1 + α(At−1 + Ft − 1) (8)
Ft = αAt−1 + (1 − α)At−2 + (1 − α)2 Ft−2 (9)
Ft = αAt−1 + (1 − α)At−2 + (1 − α)2 At−3 + (1 − α)3 Ft−3 (10)

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 33 / 44
 Time Series Models
Exponential smoothing:
Why use exponential smoothing?
Uses less storage space for data
More accurate
Easy to understand
Little calculation complexity
The smoothing constant α expresses how much our forecast will react to observed
differences.
If α is low: there is little reaction to differences.
If α is high: there is a lot of reaction to differences.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 34 / 44
 Time Series Models
Exponential smoothing:
Selecting Smoothening Constant (α):
The appropriate value of the smoothing constant, α, however, can make the
difference between an accurate forecast and an inaccurate forecast.
In picking a value for the smoothing constant, the objective is to obtain the most
accurate forecast.
Several values of the smoothing constant may be tried, and the one with the lowest
MAD could be selected.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 35 / 44
 Time Series Models
Exponential smoothing: Example

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 36 / 44
 Time Series Models
Exponential smoothing: Selecting smoothing constant.

The smoothing constant with less MAD should be selected, thus α = 0.1
Kinfe F. Weldetnsae (Samara University, College of Engineering
Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 37 / 44
 Selecting the Right Forecasting Model
Selecting Smoothening Constant (α)
1 The amount and type of available data
Some methods require more data than others
2 Degree of accuracy required
Increasing accuracy means more data
3 Length of forecast horizon
Different models for 3 month vs. 10 years
4 Presence of data patterns

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 38 / 44
 Time Series Models
Forecasting during product life cycle

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 39 / 44
 Measuring Forecast Error
1 Forecasts are never perfect
2 Need to know how much we should rely on our chosen forecasting method.
3 Measuring forecast error:
Et = At − Ft (11)
Note that:
Over-forecasts = negative errors
Under-forecasts = positive errors.
Large values of negative or positive errors shows there is bias in the forecast.

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 40 / 44
 Measures of Forecast Error
1 Mean Absolute Deviation (MAD)
Measures the total error in a forecast without regard to sign
2 Cumulative Forecast Error (CFE)
Also called running sum of forecast error (RSFE)
Measures any bias in the forecast
3 Mean Square Error (MSE)
Penalizes larger errors

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 41 / 44
 Measures of Forecast Error
Mean Absolute Deviation (MAD)
Pn
1 |At − Ft |
MAD = (12)
n
Mean Square Error (MSE)
Pn
− Ft )2 1 (At
(MSE ) = (13)
n
Cumulative Forecast Error (CFE) or running sum of forecast error (RSFE)
n
X
(CFE ) = (RSFE ) = (At − Ft ) (14)
1

N.B-Ideal values = 0 (i.e., no forecasting error)

Kinfe F. Weldetnsae (Samara University, College of Engineering
Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 42 / 44
 Measuring Accuracy: Tracking signal
The tracking signal is a measure of how often our estimations have been above or
below the actual value. It is used to decide when to re-evaluate using a model.
CFE
TS = (15)
MAD

Positive tracking signal: most of the time actual values are above our forecasted
values
Negative tracking signal: most of the time actual values are below our forecasted
values
Usually 3 ≤ TS ≥ 8, out of this range investigate!

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 43 / 44
 Measuring Forecast Accuracy and Error

Kinfe F. Weldetnsae (Samara University, College of Engineering

Industrial
andManagement
Technology,and
Department
Engineering
of Mechanical
Economics Engineering) April 2022 44 / 44