Operations Management
MATHEMATICAL
MODELS
USED IN
DECISION MAKNG
Productivity
Productivity is a relationship between the
output (product/services) and the
input (resources consumed in providing
them) of a business system.
Productivity (P) =
Output
Input
Input resources may be Wages, Cost of
Equipment, overhead costs etc.
2
Productivity Measures
Partial Measures:
Multifactor Measures:
A ratio of outputs to only one input (e.g.: labor
productivity, machine utilization, energy
efficiency)
output/(single input)
A ratio of outputs to several, but not all, inputs
output/(multiple inputs)
Total Productivity Measures:
The ratio of outputs to all inputs
output/(total inputs)
Productivity Measures
Partial measures of productivity =
Output or Output or Output or Output
Labor
Capital Materials Energy
Multifactor measures of productivity =
Output
.
Output
.
OR
Labor + Capital + Energy
Labor + Capital + Materials
Total Productivity measures =
Output
.
Labor + Capital + Energy + Materials
4
Partial (Labor) Productivity
600 Insurance policies were processed by 3
employees working 8 hours per day, 5 days in a
week
Productivity =
Policies Processed
Employee Hours
600 policies
(3 Employees)(40 Hours/Employee)
5 policies/hour
5
Multifactor Productivity
Convert all inputs & outputs to $ value
400 units of a product were made by a team of
workers at a cost of $10 each. The actual cost
inputs were:$400 labour, $1,000 materials, and
$300 overhead costs
Productivity =
Quantity at selling price
Labour + Material + O/H
=
(400 units)($10/unit)
$400 + $1000 + $300
=
$4,000/$1,700
=
2.35
Interpreting Productivity
Measures
Is the productivity measure of 2.35 in
the previous example good or bad?
Cant tell without a reference point
Compare to previous measures (e.g.:
last week) or to another benchmark
Productivity Growth Rate
Can be used to compare a process
productivity at a given time (P2) to the
same process productivity at an
earlier time (P1)
Growth Rate
P2
P1
P1
8
Productivity Growth Rate
Example:
Last week a company produced 150 units using 200 hours of labor
This week, the same company produced 180 units using 250 hours of
labor
P1
P2
150 units
200 hours
0.75 units / hour
180 units
250 hours
Growth Rate
0.72 units / hour
P2
P1
P1
0.72 0.75
0.75
0.04
or a negative 4% growth rate
9
Break-Even Analysis
Break-even analysis is used to compare
processes by finding the volume at which
two different processes have equal total
costs.
Break-even point is the volume at which
total revenues equal total costs.
Variable costs (c) are costs that vary
directly with the volume of output.
Fixed costs (F) are those costs that remain
constant with changes in output level.
10
Break-Even Analysis
Q is the volume of customers or units,
c is the unit variable cost, F is fixed
costs and p is the revenue per unit
cQ is the total variable cost.
Total cost = F + cQ
Total revenue = pQ
Break-even is where pQ = F + cQ
(Total revenue = Total cost)
11
Break-Even Analysis can tell you
If a forecast sales volume is sufficient
to break even (no profit or no loss)
How low variable cost per unit must be
to break even given current prices and
sales forecast.
How low the fixed cost need to be to
break even.
How price levels affect the break-even
volume.
12
Hospital Example
A hospital is considering a new procedure to be offered
at $200 per patient. The fixed cost per year would be
$100,000, with total variable costs of $100 per patient.
What is the break-even quantity for this service?
Q = F / (p - c) = 100,000 / (200-100) = 1,000 patients
13
Hospital Example
continued
Dollars (in thousands)
400
300
200
Quantity
(patients)
(Q)
Total Annual
Cost ($)
(100,000 + 100Q)
Total Annual
Revenue ($)
(200Q)
0
2000
100,000
300,000
0
400,000
100
|
500
1000
1500
Patients (Q)
|
2000
14
Quantity
(patients)
(Q)
Total Annual
Cost ($)
(100,000 + 100Q)
Total Annual
Revenue ($)
(200Q)
0
2000
100,000
300,000
0
400,000
(2000, 400)
Dollars (in thousands)
400
300
Total annual revenues
200
100
|
500
Quantity
(patients)
(Q)
Total Annual
Cost ($)
(100,000 + 100Q)
Total Annual
Revenue ($)
(200Q)
0
2000
100,000
300,000
0
400,000
1000
1500
Patients (Q)
|
2000
15
Quantity
(patients)
(Q)
Total Annual
Cost ($)
(100,000 + 100Q)
Total Annual
Revenue ($)
(200Q)
0
2000
100,000
300,000
0
400,000
(2000, 400)
Dollars (in thousands)
400
300
Total annual revenues
(2000, 300)
Total annual costs
200
100
Fixed costs
|
500
1000
1500
Patients (Q)
|
2000
16
Quantity
(patients)
(Q)
Total Annual
Cost ($)
(100,000 + 100Q)
Total Annual
Revenue ($)
(200Q)
0
2000
100,000
300,000
0
400,000
(2000, 400)
400
Dollars (in thousands)
Profits
300
Total annual revenues
(2000, 300)
Total annual costs
200
Break-even quantity
100
Fixed costs
Loss
0
|
500
1000
1500
Patients (Q)
|
2000
17
Two Processes and
Make-or-Buy Decisions
Breakeven analysis can be used to choose
between two processes or between an
internal process and buying those services
or materials.
The solution finds the point at which the total
costs of each of the two alternatives are
equal.
The forecast volume is then applied to see
which alternative has the lowest cost for that
volume.
18
Breakeven for
Two Processes
19
Q=
Breakeven for
Two Processes
Q=
Fm Fb
cb cm
12,000 2,400
2.0 1.5
20
Q=
Breakeven for
Two Processes
Fm Fb
cb cm
Q = 19,200 salads
21
Under Certainty
22
Under Certainty
Alternative
Small facility
Large facility
Do nothing
Possible
Future Demand
Low
High
200
160
0
270
800
0
Example A.5
23
Under Certainty
Alternative
Small facility
Large facility
Do nothing
Possible
Future Demand
Low
High
200
160
0
270
800
0
If future demand will be low
Example A.5
24
Under Certainty
Alternative
Small facility
Large facility
Do nothing
Possible
Future Demand
Low
High
200
160
0
270
800
0
If future demand will be low Choose the small facility.
Example A.5
25
Under Certainty
Alternative
Small facility
Large facility
Do nothing
Possible
Future Demand
Low
High
200
160
0
270
800
0
If future demand will be low Choose the small facility.
Example A.5
26
Under Uncertainty
Alternative
Small facility
Large facility
Do nothing
Example A.6
Possible
Future Demand
Low
High
200
160
0
270
800
0
27
Under Uncertainty
Alternative
Small facility
Large facility
Do nothing
Example A.6
Possible
Future Demand
Low
High
200
160
0
270
800
0
Maximin
28
Under Uncertainty
Alternative
Small facility
Large facility
Do nothing
Example A.6
Possible
Future Demand
Low
High
200
160
0
270 Maximin
800
0 Best of the worst
29
Under Uncertainty
Alternative
Small facility
Large facility
Do nothing
Example A.6
Possible
Future Demand
Low
High
200
160
0
270 Maximin Small
800
0 Best of the worst
30
Under Uncertainty
Alternative
Small facility
Large facility
Do nothing
Example A.6
Possible
Future Demand
Low
High
200
160
0
270
800
0
Maximin Small
Maximax
31
Under Uncertainty
Alternative
Small facility
Large facility
Do nothing
Possible
Future Demand
Low
High
200
160
0
270
800
0
Maximin Small
Maximax
Best of the best
Example A.6
32
Under Uncertainty
Alternative
Small facility
Large facility
Do nothing
Possible
Future Demand
Low
High
200
160
0
270
800
0
Maximin Small
Maximax Large
Best of the best
Example A.6
33
Under Uncertainty
Alternative
Small facility
Large facility
Do nothing
Example A.6
Possible
Future Demand
Low
High
200
160
0
270
800
0
Maximin Small
Maximax Large
Laplace
34
Under Uncertainty
Alternative
Small facility
Large facility
Do nothing
Small facility
Example A.6
Possible
Future Demand
Low
High
200
160
0
270
800
0
Maximin Small
Maximax Large
Laplace
0.5(200) + 0.5(270) = 235
Best
weighted
payoff
35
Under Uncertainty
Alternative
Small facility
Large facility
Do nothing
Small facility
Large facility
Example A.6
Possible
Future Demand
Low
High
200
160
0
270
800
0
Maximin Small
Maximax Large
Laplace
0.5(200) + 0.5(270) = 235
0.5(160) + 0.5(800) = 480
Best
weighted
payoff
36
Under Uncertainty
Alternative
Small facility
Large facility
Do nothing
Small facility
Large facility
Example A.6
Possible
Future Demand
Low
High
200
160
0
270
800
0
Maximin Small
Maximax Large
Laplace Large
0.5(200) + 0.5(270) = 235
0.5(160) + 0.5(800) = 480
Best
weighted
payoff
37
Under Uncertainty
Alternative
Small facility
Large facility
Do nothing
Example A.6
Possible
Future Demand
Low
High
200
160
0
270
800
0
Maximin Small
Maximax Large
Laplace Large
Minimax Regret
38
Under Uncertainty
Possible
Future Demand
Low
High
Alternative
Small facility
Large facility
Do nothing
200
160
0
270
800
0
Maximin Small
Maximax Large
Laplace Large
Minimax Regret
Regret
Low Demand
High Demand
Small facility
Example A.6
200 200 = 0
800 270 = 530
Best
worst
regret
39
Under Uncertainty
Possible
Future Demand
Low
High
Alternative
Small facility
Large facility
Do nothing
200
160
0
270
800
0
Maximin Small
Maximax Large
Laplace Large
Minimax Regret
Regret
Low Demand
High Demand
Small facility
Large facility
Example A.6
200 200 = 0
200 160 = 40
800 270 = 530
800 800 = 0
Best
worst
regret
40
Under Uncertainty
Possible
Future Demand
Low
High
Alternative
Small facility
Large facility
Do nothing
200
160
0
270
800
0
Maximin Small
Maximax Large
Laplace Large
Minimax Regret Large
Regret
Low Demand
High Demand
Small facility
Large facility
Example A.6
200 200 = 0
200 160 = 40
800 270 = 530
800 800 = 0
Best
worst
regret
41
Under Uncertainty
Alternative
Small facility
Large facility
Do nothing
Example A.6
Possible
Future Demand
Low
High
200
160
0
270
800
0
Maximin Small
Maximax Large
Laplace Large
Minimax Regret Large
42
Decision Trees
Decision Trees are schematic
models of alternatives available along
with their possible consequences.
They are used in sequential decision
situations.
Decision points are represented by
squares.
Event points are represented by
circles.
43
Decision Trees
After drawing a decision tree, we solve it by working
from right to left, starting with decisions farthest to the
right, and calculating the expected payoff for each of
its possible paths.
We pick the alternative for that decision that has the
best expected payoff.
We saw off, or prune, the branches not chosen by
marking two short lines through them.
The decision nodes expected payoff is the one
associated with the single remaining branch.
44
ANALYZING A DECISION TREE
A retailer must decide whether to build a small or a large facility
at a new location. Demand at a location can be either small or
large with probabilities estimated to be 0.4 and 0.6 respectively.
If a small facility is built and demand proves to be high, the
manager may choose not to expand (payoff = $223,000) or to
expand (payoff = $270,000). If a small facility is built and
demand is low there is no reason to expand and the payoff is
$200,000. If a large facility is built and the demand proves to be
low, the choice is to do nothing ($40,000) or to simulate
demand through local advertising. The response to advertising
may be either modest or sizeable, with their probabilities
estimated to be 0.3 and 0.7 respectively. If it is modest, the
payoff is estimated to be only $20,000; the payoff grows to
$220,000 if the response is sizeable. Finally, if a large facility is
built and demand turns out to be high, th payoff is $800,000.
Analyze with help of a decision tree to determine the
expected payoff for each decision and event node. Then
decide which alternative has the higher expected payoff. 45
ANALYZING A DECISION TREE
1)
The decision tree is shown in the figure with
event probability and payoff for each event. The
decisions are:
To build a small or large facility
2)
To expand now or later (small facility with high
demand)
3)
Whether to advertise or not (large facility with low
demand)
Now we calculate the expected payoffs
46
Drawing the Tree
47
Drawing the Tree
continued
Low demand [0.4]
$200
Dont expand
$223
Expand
$270
48
Completed Drawing
Low demand [0.4]
$200
Dont expand
$223
Expand
$270
Do nothing
$40
Modest response [0.3]
Advertise
Sizable response [0.7]
High demand [0.6]
$800
$20
$220
49
Solving Decision #3
Low demand [0.4]
$200
Dont expand
$223
Expand
$270
Do nothing
0.3 x $20 = $6
$40
Modest response [0.3]
Advertise
Sizable response [0.7]
$6 + $154 = $160
$20
$220
0.7 x $220 = $154
High demand [0.6]
$800
50
Solving Decision #3
Low demand [0.4]
$200
Dont expand
$223
Expand
$270
Do nothing
$40
Modest response [0.3]
Advertise
$160
Sizable response [0.7]
$160
High demand [0.6]
$800
$20
$220
51
Solving Decision #2
Low demand [0.4]
$200
Dont expand
$223
Expand
$270
1
$270
Expanding has a
higher value.
Do nothing
$40
Modest response [0.3]
Advertise
$160
Sizable response [0.7]
$160
High demand [0.6]
$800
$20
$220
52
Solving Decision #1
Low demand [0.4]
$200 x 0.4 = $80
$242
$242
Dont expand
$223
Expand
$270 x 0.6 = $162
$270
1
Do nothing
$40
Modest response [0.3]
Advertise
$160
Sizable response [0.7]
$160
High demand [0.6]
$800
$20
$220
53
Solving Decision #1
Low demand [0.4]
$200
$242
Dont expand
$223
Expand
$270
1
$270
Do nothing
$40
Advertise
$160
0.4 x $160 = $64
$544
Modest response [0.3]
Sizable response [0.7]
$160
High demand [0.6]
$800 x 0.6 = $480
$20
$220
54
Solving Decision #1
Low demand [0.4]
$200
$242
Dont expand
$223
Expand
$270
1
$270
Do nothing
$40
$544
Advertise
$160
Sizable response [0.7]
$160
$544
Modest response [0.3]
$20
$220
High demand [0.6]
$800
55
ANALYZING A DECISION TREE
The best alternative thus is to
build the large facility.
Subsequent decisions can be
made after learning whether
demand is actually low or
high.
56
