Decision Making

Dr Anisur Rahman

Decision Making
Decision making is the cognitive process leading
to the selection of a course of action among
variations.
Every decision making process produces a final
choice.
Decision making is a reasoning process which
can be rational or irrational, can be based on
explicit assumptions.

Decision Making Tools

Allocation

or Assignment Human
resource planning;

Equipment

planning;

Productivity

comparison;

Investment;
Repair

replacement Strategy

Human Resources Planning

Purpose:

To maintain an adequate and uniform supply of

suitable experienced labour.
To reduce overmanning during downturns in
demand
To avoid sharp fluctuation in each trade.

Factors Influencing Demand & Supply of Labour

Wages & Rewards -Incentive
Skills
Other factors such as Geographical location, working conditions,
Social

Allocation/Assignment Method
Say you have three rushed tasks (T1), (T2) &
(T3) and you have ADAMS, BROWN and
COOPER are available to perform tasks at
different speed/cost).
T1

T2

T3

ADAMS

$11

$14

$6

BROWN

$8

$10

$11

COOPER

$9

$12

$7

Assumptions:
1. No. of People (rows) = No. Tasks (columns)
2. The numbers within the matrix are the costs
associated with each particular task.
Six alternatives are available

3! = 3(2)(1) = 6.

Step 1: Row operation select the lowest cost element in each row and
subtract this element from all elements in that row to develop a new matrix

ADAMS
BROWN

T1
$11
$8

T2
$14
$10

T3
$6
$11

COOPER

$9

$12

$7

ADAMS
BROWN

T1
$5
$0

T2
$8
$2

T3
$0
$3

COOPER

$2

$5

$0

Check each row and

column has at least
one 0; if not go step 2

Step 2: Column operation select the lowest cost element in each row in
the new matrix and subtract this element from all elements in that row to
develop a new matrix

ADAMS
BROWN

T1
$5
$0

T2
$6
$0

T3
$0
$3

COOPER

$2

$3

$0

Step 3: Strike off all the columns and rows that contain at least one zero
with minimum number of horizontal and vertical lines

ADAMS
BROWN

T1
$5
$0

T2
$6
$0

T3
$0
$3

COOPER

$2

$3

$0

Step 4: Check: # of lines = # of jobs, if not means optimisation has not yet
arrived. In that case select the minimum uncovered element and subtract
this min element from all the uncovered elements to get a new matrix
And strike off all the rows and columns again with minimum numbers of
horizontal and vertical lines

ADAMS
BROWN

T1
$3
$0

T2
$4
$0

T3
$0
$3

COOPER

$0

$1

$0

Step 4: Check #Lines = #jobs, if yes optimal solution arrived

Step 5: Select the row/column with minumum # of zero, assign the row Person to
the column Jobs corresponding to the zero and strike off that column and row

ADAMS
BROWN

T1
$3
$0

T2
$4
$0

T3
$0
$3

COOPER

$0

$1

$0

Assign Job
T3 to Adam

Step 5: Do the same things for remaining rows and columns and assign the
corresponding Persons to Jobs

ADAMS
BROWN

T1

T2

$0

$0

COOPER

$0

$1

ADAMS
BROWN

T1

T2
$0

COOPER

Total cost = 6+10+9 = $25

T3

T3

Assign Job
T1 to Cooper

Assign Job
T2 to Brown

The assignment is made as follows:

1. Select the row or column with fewest number of
zeroes in it
2. Select an element in that row/column that is zero
3. Assign the person in that row to that job in the
column
4. Delete assigned row and column
5. Repeat the preceding procedure.

You can have an optimal solution by Maximising efficiency

Example:
The only difference is that you need to convert the matrix first
into a minimising opportunity cost table and then proceed with
the previous four steps as before.

Efficiency

Opportunity Cost

T1

T2

T3

T4

ADAMS

20

60

50

55

BROWN

60

30

80

75

COOPER

80

100

90

80

DAVIS

65

80

75

70

T1

T2

T3

T4

ADAMS

80

40

50

45

BROWN

40

70

20

25

COOPER

20

10

20

DAVIS

35

20

25

30

ADAMS
BROWN
COOPER
DAVIS15

ADAMS
BROWN
COOPER
DAVIS 0

T1
40
20
20

T2
0
50
0
0

T1
25
5
5

T3
10
0
10
5

T2
0
50
0
0

T4
5
5
20
10

T3
10
0
10
5

T4
0
0
15
5

4 Straight lines covering all zeros in the matrix

Optimal Solution:
Cooper will definitely perform Task T2
(100%)
Davis will perform Task T1
(65%)
Brown Task 3 (80%) & finally Adams will perform Task 4 (55%)

Matrix Reduction Question

Your company has one surplus truck in each of the towns A, B, C,
D, and E and one deficit truck in each of the construction sites 1, 2,
3, 4, 5 and 6. The distance between the towns and the construction
sites in kilometres is shown in the matrix below (see Table 1).
Assign the trucks from towns to construction sites to minimise the
total distance covered.
Site 1

Site 2

Site 3

Site 4

Site 5

Site 6

12

10

15

22

18

10

18

25

15

16

12

11

10

14

10

13

13

12

12

11

13

10

Equipment Planning
You need to buy an equipment
Age at
Purchase Price (k)

Age of
machine

Resale
price (k)

Maintenance
(k)

40

31

28

20

18

11

10

Determine what age of machine should be purchased and when

it should be replaced.

Maintenance
Cost

Ownership Cost
Age at Purchase

Age at Purchase
0
1
2
3
(40) (31) (20) (11)

0
1
2
3
(40) (31) (20) (11)

1(28)

2 (18)

$30 $21 $10

3 (10)

13

10

$36 $27 $16

$7

4 (4)

22

19

15

1(28)

$12

2 (18)

$22 $13

3 (10)
4 (4)

Yearly
Total Cost

Age at Purchase
0
1
2
3
(40) (31) (20) (11)

1(28)

15

2 (18)

14.5

17

3 (10)

14.3 15.5

16

4 (4)

14.5 15.3 15.5 16

Comparing Productivity
Two equipment A & B are being considered for aggregate
production. (Price: $.25 per kg).
Annual demand: 500,000 kg per year
Equipment A costs $150,000 to buy. It has an expected life of 10
years (If it works 12 hrs/day). Annual expenses (maintenance and
operational costs) amount to $100,000. Equipment A has a
production rate of 250 kg/hr.
Equipment B is highly automated and costs $240,000 and has an
expected life of 12 years (If it works 8 hrs/day). Annual expenses
for B amount to $75,000. B has a production rate of 300 kg/hr.
Assume 250 working days / year.

Equipment A
It requires 500,000/(250) = 2,000 hrs of production/year
Operating life = 10 years working 12 hrs day
= 12 (250) = 3,000 hrs / yr.
Life expected at the 2,000 hrs/year rate = 10(3,000)/2,000 = 15 years
Income = 500,000 (0.25)
= $125,000 per year
less expenses
= $100,000
Profit (savings)
= $25,000
So you need to invest $150,000 to get $25,000 per year for 15 years
P =25000x15/150000 = 2.5

Equipment B
It requires 500,000/ (300) = 1,666 hrs of production/year
Operating life = 12 years working 8hrs day
= 8 (250) = 2,000 hrs / yr.
Life expected at the 1,666 hrs/year rate = 12(2,000)/1,666 = 14.5years
Income = 500,000 (0.25)
less expenses
Profit

= $125,000
= $ 75,000
= $ 50,000

You need to invest $240,000 to get $50,000 per year for 14.5 years
P = 50000x14.5/240000 = 3.02
Therefore, B is the best proposition.

Investment decisions

If you could somehow determine precisely what would happen as a result of

choosing each option in a decision, making business decisions would be easy. You
could simply calculate the value of each competing option and select the one with
the highest value.

In the real world, decisions are not quite this simple. However, the process of
decision-making still requires choosing the most valuable option--most valuable
being, in this case, the option that has the highest Expected Monetary Value (EMV),
a measure of probabilistic value.

Suppose you are given the opportunity to play a simple game. A friend flips a coin.
If it comes up heads, you win $100. If it comes up tails, you win nothing. What is the
value of this game to you? Stated another way, how much would you pay to play
this game?

Each time you play the game you have a 50% chance of winning $100 and a 50%
chance of winning nothing. If you were to play the game many times, on average
you would win $50 for every time you played. Therefore, $50 is the EMV for this
game.

The EMV is calculated by multiplying each outcome value by its probability and
adding all of the results together.

For the coin example EMV can be calculated by using the equation

EMV = $100 * 0.50 + $0 * 0.50 = $50

Decision Making under Risk/Uncertainty

Undertake a project A
Undertaking project B
Undertaking nothing

Boom
200,000
100,000
0

Recession
180,000
20,000
0

This is highly dependent on the probability of having a

recession, say it is 50/50 for this example
For each alternative, calculate the Expected Monetary Value
(EMV)
EMV (A) = 0.5(200,000) 0.5 (180,000) = $10,000
EMV (B) = 0.5(100,000) 0.5 (20,000) = $40,000
EMV (Nothing) = 0
Select the option with the maximum (EMV) Take the project B.

Opportunity Loss
we can achieve the same answer, if we create an opportunity loss table
(how much will it cost me if I do not choose the best alternative)
Undertake a project A
Undertaking project B
Undertaking nothing

Boom
200,000
100,000
0

Recession
180,000
20,000
0

Opportunity Loss Table

Boom
Recession
Undertake project A
200,000-200,000
0-(180,000)
Undertake project B
200,000-100,000
0-(20,000)
Undertaking nothing
200,000 0
0-( 0)
Expected Opportunity Loss (A) = 0.5(0) + 0.5(180,000) = $90,000
EOL (B)= 0.5(100,000)+ 0.5(20,000)= $60,000
EOL (nothing) = 0.5(200,000) + 0.5(0) = $100,000

Suppose you can be sure of the market status (boom/recession), if

you conduct a market analysis
This will give you a perfect answer, however it will cost you
How much You are ready to spend to get this perfect answer?
The following two terms will help you make the right decision
The Expected value with Perfect information (EP) =
(best outcome for boom/recess.)* (probability of its occurrence)
&
The Expected Value of Perfect Information (EVPI)
EVPI = EP Max. EMV

Boom
200,000
100,000
0

Undertake project A
Undertake project B
Undertaking nothing
Maximum EMV = $40,000
The best outcome Boom
The best outcome for Recession

= $200,000
= $0

The expected value with perfect information (EP)=

(200,000)(0.5) + (0) (0.5) = $100,000
EVPI = 100,000 - 40,000 = $60,000 = Minimum EOL
So Do not pay more than 60,000 for the PERFECT information

Recession
180,000
20,000
0

D.M. under Uncertainty

When the probability of occurrence can be assessed,
Max. EMV or Min. EOL can be used, but what happens if you cannot
assess the outcome probability.
In our example, we assumed 50/50 chance... What happens if we have
no clue about the situation? This is Uncertainty.

Because I have no clue, you can be a gambler (optimistic) max. profit

or conservative (pessimistic) min. loss.
Undertake project A
Undertake project B
Undertaking nothing

Boom
200,000
100,000
0

Recession
180,000
20,000
0

Or You can simply go for the highest average outcome

Average!!! (sound like 50/50 again).
Undertake project A
Undertake project B
Undertaking nothing

Boom
200,000
100,000
0

Recession Av./2
180,000
10 k
20,000
40 k
0
0

Or finally, go for the minimum opportunity Loss

Boom
Recession
Undertake project A
0
180,000
Undertake project B
100,000
20,000
Undertaking nothing
200,000
0
Go for the alternative that minimises the opportunity Loss

Max.
180k
100k
200k

Replacement Strategy
A set of 100 lamps is used to give warning of a series of road works.
Each lamp is powered by a battery which is expected to fail
sometime during a period of 3 months. Replacement of of failed
batteries may take place only at the end of the month (to replace
each failed battery it costs $10.
You may choose to replace the whole set for $500 (whether failed
or not). The following table gives data relating to the survival of
the batteries:
Month of Use
Probability of failure
during this month

1
0.2

2
0.5

3
0.3

100

80
20

100

1
0.2

30
50

30

month 2
30

month 1
80

2
0.5
month 3

3
0.3
month 4

30

50

40

20

10
16

2
6
8

10

1
Failure per month

20

54

51

43

Failure per month

20
Cost of replacement 200
Cumulative failures 20

54
540
74

51
510
125

43
430
168