Examiners’ commentaries 2023

Examiners’ commentaries 2023

(May)

MN2032 Management science methods

Important note

This commentary reflects the examination and assessment arrangements for this
course in the academic year 2022–2023. The format and structure of the
examination may change in future years, and any such changes will be publicised
on the virtual learning environment (VLE).
Information about the subject guide and the Essential reading references
Unless otherwise stated, all cross-references will be to the latest version of the
subject guide (2018). You should always attempt to use the most recent edition
of any Essential reading textbook, even if the commentary and/or online reading
list and/or subject guide refer to an earlier edition. If different editions of Essential
reading are listed, please check the VLE for reading supplements – if none are
available, please use the contents list and index of the new edition to find the
relevant section.

General remarks

Learning outcomes
At the end of this course and having completed the Essential reading and Activities,
you should be able to:
• discuss the main techniques and problem-structuring methods used within
management science
• critically appraise the strengths and limitations of these techniques and problem-
structuring methods
• carry out simple exercises using such techniques and problem-structuring
methods themselves (or explain how they should be done)
• commission more advanced exercises.

What the examiners are looking for

You need to be clear that, aside from being able to display knowledge of the subject
guide (both in qualitative and in quantitative questions), you are expected to apply
that knowledge, perhaps to examination questions not encountered in precisely the
same format before.
Some questions may contain features that have not been directly examined
previously. It is disappointing to see how many candidates are completely unable to

1
MN2032 Management science methods

cope with these features. You are reminded that success in an examination is not
simply a matter of practising all previous examination questions in the hope that
similar questions will recur. Rather, the examiners expect you to have a clear
understanding of the principles underlying both quantitative and qualitative
approaches and to be able to apply them appropriately. This may be in a question
that is similar in form to a previous examination question, or it may be in a question
that contains features that have not been encountered before.
Some candidates appear to adopt the strategy for qualitative questions of simply
memorising significant quantities of text from the subject guide in the hope that
reproducing that text in the examination will lead to success. You are reminded that
the examiners are looking for you to show understanding of the material presented in
the subject guide. This may entail expressing concepts and ideas in your own words.
With regard to qualitative questions in particular, learning an answer to a previously
asked question and then repeating it in the hope that it will contain sufficient points to
address an unfamiliar question, will not be a successful strategy. You need to focus
on the question asked, not hope that an unfocused answer will be rewarded by the
examiners.
You should be clear that for all questions, but especially for quantitative questions,
answers need to be supported by appropriate calculation/written reasoning. For
example, a statement may be given by the examiners in the question and you have
to agree or disagree with the statement. Candidates who give a correct answer, but
without an appropriate reason, will not score highly. A question can be drawn from
any topic in the subject guide. In particular, a question may mix both qualitative and
quantitative topics.
An issue for a number of candidates in the examination in 2023 was a failure to read
the question in sufficient detail and understand the requirements of the examiner. In
particular a number of questions below require the student to copy a table into their
answer and fill in the values required. In doing so typically the required accuracy is
CLEARLY STATED in the details of the table given in the question (e.g. % 2dp, or
4dp, or % integer, or £’m 2dp). Candidates need to be aware therefore that an answer
to the accuracy stated is required.

2
 Examiners’ commentaries 2023

Examination revision strategy

Many candidates are disappointed to find that their examination performance is

poorer than they expected. This can be due to a number of different reasons and
the Examiners’ commentaries suggest ways of addressing common problems
and improving your performance. We want to draw your attention to one particular
failing – ‘question spotting’, that is, confining your examination preparation to a
few question topics which have come up in past papers for the course. This can
have very serious consequences.
We recognise that candidates may not cover all topics in the syllabus in the same
depth, but you need to be aware that examiners are free to set questions on any
aspect of the syllabus. This means that you need to study enough of the syllabus
to enable you to answer the required number of examination questions.
The syllabus can be found in the Course information sheet in the section of the
VLE dedicated to this course. You should read the syllabus very carefully and
ensure that you cover sufficient material in preparation for the examination.
Examiners will vary the topics and questions from year to year and may well set
questions that have not appeared in past papers – every topic on the syllabus is
a legitimate examination target. So although past papers can be helpful in
revision, you cannot assume that topics or specific questions that have come up
in past examinations will occur again.
If you rely on a question spotting strategy, it is likely you will find yourself
in difficulties when you sit the examination paper. We strongly advise you
not to adopt this strategy.

3
MN2032 Management science methods

Examiners’ commentaries 2023

(May)

MN2032 Management science methods

Important note
This commentary reflects the examination and assessment arrangements for this
course in the academic year 2022–2023. The format and structure of the
examination may change in future years, and any such changes will be publicised
on the virtual learning environment (VLE).
Information about the subject guide and the Essential reading references
Unless otherwise stated, all cross-references will be to the latest version of the
subject guide (2018). You should always attempt to use the most recent edition
of any Essential reading textbook, even if the commentary and/or online reading
list and/or subject guide refer to an earlier edition. If different editions of Essential
reading are listed, please check the VLE for reading supplements – if none are
available, please use the contents list and index of the new edition to find the
relevant section.

Comments on specific questions

Candidates should answer FOUR of the following EIGHT questions. All
questions carry equal marks. If you answer more than four questions ONLY the
first four will be marked.
Question 1
Apply SSM (Soft Systems Methodology) to two problems with which you are
familiar.
Your answer to this question for each problem that you consider should be
limited to 5 pages, so a maximum of 10 pages in total.
Reading for this question
This question relates to Chapter 2 of the subject guide.
Approaching the question
Points that the examiners would expect to see here (for each of the two problems
considered) are:
• A clear statement (in words) of the problem considered, not just a single phrase.
• Appropriate application of SSM to the problem:
o root definition (statement of the ideal)
o CATWOE for the root definition
o explicit and clear check of root definition by CATWOE
o application of the stages to the problem

4
 Examiners’ commentaries 2023

Question 2
A company is considering using Markov theory to analyse consumers
switching between three different brands of microwave instant meals. An
analysis of consumer behaviour data has produced the transition matrix shown
below for the probability of switching every two months between brands. The
current market shares for the three brands are 54%, 19% and 27% for brands
A, B and C respectively.
To brand
A B C
A 0.91 0.07 0.02
From
brand B 0.71 0.08 0.21
C 0.63 0.26 0.11

Here, for example, there is a probability of 0.02 that a customer switches from
brand A to brand C.
a) Copy the following table and fill in the market shares for these brands, after
two months, and in the long-run.

Brand Brand Brand

A B C

After two months (%,

2dp)

Long-run (%, 2dp)

b) The company thinks that the transition probabilities seen in part (a) above
are no longer indicative of customer preference. They have commissioned
a market survey of customers which has resulted in the table shown below.
The researchers for this market survey have asked 1000 brand A
customers, 1500 brand B customers and 1500 brand C customers as to their
preference in switching to a different brand. So in the table below 123
customers said they would switch from brand B to brand C.

From brand
A B C
A 75 637 748
To brand B 505 740 52
C 420 123 700

Copy the following table and fill in the market shares for these brands in the
long-run.

5
MN2032 Management science methods

Brand A Brand B Brand C

Long-run (%, 2dp)

Reading for this question

This question relates to Chapter 6 of the subject guide.
Approaching the question
a) For the market share after two months (so one transition) we take the initial
system state and matrix multiply by the transition matrix. This gives

Brand A Brand B Brand C

After two months (%, 2dp) 79.64 12.32 8.04

For the long-run prediction we need to set up [x1,x2,x3] where

[x1,x2,x3] = [x1,x2,x3](transition matrix) and x1 + x2 + x3 = 1. Expanding we get
x1 = 0.91x1 + 0.71x2 + 0.63x3
x2 = 0.07x1 + 0.08x2 + 0.26x3
x3 = 0.02x1 + 0.21x2 + 0.11x3
x1 + x2 + x3 = 1
and solving these equations simultaneously

As fractions the answers are:

x1 = 0.8837
x2 = 0.0781
x3 = 0.0383

The required table is:

Brand A Brand B Brand C

Long-run (%, 2dp) 88.37 7.81 3.83

b) Here we need to first find the new transition probabilities. We divide each
element in the matrix given in the question by its column sum. However as the
matrix is given in a To-From form we need to transpose it to get the transition
probabilities. These are|:

To brand
A B C
A 0.0750 0.5050 0.4200
From brand B 0.4247 0.4933 0.0820
C 0.4987 0.0347 0.4667

For the long-run prediction we need to set up [x1,x2,x3] where

[x1,x2,x3] = [x1,x2,x3](transition matrix) and x1 + x2 + x3 = 1. Expanding we get
x1 = 0.0750x1 + 0.4247x2 + 0.4987x3
x2 = 0.5050x1 + 0.4933x2 + 0.0347x3
x3 = 0.4200x1 + 0.0820x2 + 0.4667x3
x1 + x2 + x3 = 1
and solving these equations simultaneously

6
 Examiners’ commentaries 2023

As fractions the answers are:

x1 = 0.3320
x2 = 0.3525
x3 = 0.3156

The required table is:

Brand A Brand B Brand C

Long-run (%, 2dp) 33.20 35.25 31.56

Question 3
a) Solve the following linear program graphically using an iso-cost/iso-profit
line. The vertices (corners) of the feasible region should be clearly labelled
A, B, C etc.

Minimise z = 4x – 2y
Subject to 2x + 3y ≤ 12 Constraint 1
x + 3y ≤ 8 Constraint 2
2x – 4y ≥ 1 Constraint 3
x,y ≥ 0

Copy the tables shown below and fill in the values required.

Optimal value (4dp) Opportunity cost (4dp)

x
y
z

Shadow price (4dp)

Constraint 1
Constraint 2
Constraint 3

b) If constraint 3 is now removed (so the problem simply has two variables
with Constraint 1 and Constraint 2) then indicate the new feasible region by
labelling its corners using M, N, O etc. Copy the table and fill in the values
required.

7
MN2032 Management science methods

Optimal value (4dp)

x
y
z

Reading for this question

This question relates to Chapter 8 of the subject guide.
Approaching the question
a) The LP diagram is:

with the optimal solution being at A, namely:

Optimal value (4dp)
x 0.5
y 0
z 2

For x the opportunity cost (reduced cost) is zero (or undefined) since it is
already > 0.

For y we set y=1 and resolve to get the reduced cost.

For constraints 1 and 2 the shadow price is zero as they are slack.

For constraint 3, the shadow price is given by changing the right-hand side by
one (say to 2) and resolving.

So we have:

8
 Examiners’ commentaries 2023

Opportunity cost (4dp)

x 0 (or undefined)
y 6

Constraint Shadow price (4dp)

1 0
2 0
3 2

b) The constraint removal will alter the feasible region, as below

with the optimal solution being at N, namely:

Optimal value (4dp)

x 0
y 2.6667
z -5.3333

Question 4
You have been given the pairwise comparison matrix for three objectives
shown below.

Objective
1 2 3
1 1 2 6
Objective 2 - 1 7
3 - - 1

9
MN2032 Management science methods

Here, for example, the judgement as to objective 1 as compared with objective

3 is rated 6.

a) Are the judgements made with respect to these objectives reasonably

consistent or not (assume the value of RI for n=3 is 0.58)?

b) The relative judgment of two possible courses of action (A and B) with

respect to the above three objectives is as shown below.

Assuming that the judgements made with respect to the three objectives
are reasonably consistent, which course of action (A or B) should be taken?

A B
A 1 3
Objective 1
B - 1

A B
A 1 -
Objective 2
B 4 1

A B
A 1 8
Objective 3
B - 1

c) Discuss how, in practice, you might use simulation to evaluate which

course of action (A or B) you might take if you are told that the judgement
as to objective 1 as compared with objective 3 is no longer certain to be
rated 6 (as in part (a)), but instead may be either 6 with probability 0.3, 7
with probability 0.5 or 8 with probability 0.2.

Reading for this question

This question relates to Chapter 10 of the subject guide.
Approaching the question
a) The details of the calculation are:

Comparison matrix Objective

1 2 3
Objective 1 1 2 6
2 0.5000 1 7
3 0.1667 0.1429 1
Column
sum 1.6667 3.1429 14.0000

Normalised matrix 1 2 3 Row average

1 0.6000 0.6364 0.4286 0.5550
2 0.3000 0.3182 0.5000 0.3727
3 0.1000 0.0455 0.0714 0.0723

10
 Examiners’ commentaries 2023

Weights 0.5550 0.3727 0.0723

Column
vector 1.7342
1.1563
0.2181

max 3.0813
CI RI CI/RI
0.0407 0.5800 0.0702

as the ratio CI/RI is less than 0.1 the judgements are consistent.

b) The calculations are:

Choice A 1.00 3.00

B 0.33 1.00
Column
sum 1.33 4.00

Row
Normalised matrix A B average
A 0.7500 0.7500 0.7500
B 0.2500 0.2500 0.2500

The calculation for the other two objectives is summarised in the table below

Objective A B
1 0.5550 0.75 0.25
2 0.3727 0.2 0.8
3 0.0723 0.8889 0.1111
0.5551 0.4449

So the choice is A.

c) For this part of the question the examiners would expect:

• mention of the use of Excel/software to evaluate/calculate AHP decision
given an objective weighting
• random objective comparison generation in line with the probability
distribution given in the question
• constructing a frequency distribution showing:
o number of times judgements inconsistent
o number of times A or B chosen IF judgements were consistent.

Question 5
a) A small shop buys a weekly magazine from a publisher for a unit price of
£5.50 and sells it to members of the public for 7.35. Any magazines left over
at the end of the week are returned to the publisher and for each magazine

11
MN2032 Management science methods

returned the shop owner receives £0.45. Past experience indicates that the
demand for this magazine from members of the public is expected to have
a normal distribution with mean 157 copies and variance 83. By making use
of the table associated with the standard normal distribution (with mean
zero and variance one) presented below what would be your advice as to
how many copies the shop owner should order from the publisher?

b) The shop owner also sells cigarettes that she orders from a supplier. She
pays her supplier £6.75 for each packet. Currently she sells 30 packets per
week. She estimates that the costs associated with making an order are
£3.70 and interest rates on the overdraft she uses for cash flow purposes
are currently 7% per annum. Her supplier offers a quantity discount such
that if she were to order 100 packets or more at a time she only pays £6.00
per packet. What advice can you offer her as to a recommended order
quantity and what would be the associated total cost per year? Assume for
this part of the question that there are 52 weeks in a year.

c) An alternative supplier of the cigarettes that she sells has approached her
and offered the packets for £6.00 per packet if she orders 175 packets each
time she orders. The shop owner estimates that the costs associated with
making an order from this alternative supplier would be £2.55. She feels that
she can reject this offer as the order quantity of 175 is higher than the order
quantity of 100 (for the same price of £6.00) as given in part (b) of the
question. Is she correct that she can reject this offer? Give your reasons,
without doing any numeric calculation.

Furthermore, this alternative supplier will reduce the price per packet to
£5.50, but only if she orders 750 packets each time she makes an order.

What advice can you offer her as to a recommended order quantity and what
would be the associated total cost per year?

Assume for this part of the question that there are 52 weeks in a year.

Reading for this question

This question relates to Chapter 5 of the subject guide.
Approaching the question
a) The equation to use is CP = Cunder/(Cunder + Cover)
Here Cunder = the cost of underestimating demand by one unit = the lost profit =
7.35-5.50 = 1.85
Cover = the cost of overestimating demand by one unit =5.50-0.45 = 5.05

Hence CP =1.85/(1.85+5.05) = 0.2681

When we have the Normal distribution with mean 157 and variance 83 the
quantity to order (x) is given by (x – 157)/√83 = the value from the N(0,1) with
0.2681 of probability to the left of it. Here the value from tables is –0.62

So (x – 157)/√83 = –0.62
x = 151.4 say 151 copies (or 152)

b) We need an EOQ calculation. The EOQ formula is EOQ = √(2Rco/ch), where

R = 30×52 =1560 and co = 3.7 and P=6.75

12
 Examiners’ commentaries 2023

ch =0.07(6.75) = 0.4725 per year

so EOQ = √(2×1560×3.7/0.4725)
= 156.3 (say 156 or 157)

As this EOQ exceeds 100, when the quantity discount applies, it would be
incorrect to cost it with a price of 6.75.

One way of expressing this is to say that this is an ‘invalid EOQ’. Costing it with
a price of 6 is also incorrect as the EOQ should be recalculated, as below.

With the quantity discount the new purchase cost is 6 and the new holding cost
is 0.07(6) = 0.42

The new EOQ is √(2×1560×3.7/0.42) = 165.8 (say 166, or 165)

This is feasible with respect to the minimum order quantity of 100 so need to
cost the order quantity of 166
Using Q=166 the cost per year is RP + (chQ/2) + (coR/Q)
=1560×6 + (0.42×166/2) + (3.7×1560/166)
= 9429.6

So advice is to order 166 at a time.

c) With regard to: She feels that she can reject this offer as the order quantity of
175 is higher than the order quantity of 100 (for the same price of £6.00) as
given in part (b) of the question.

Logically she can only reject the offer if she is absolutely sure that it would be
more costly. Without doing any numeric calculation it is not true that she
can reject the offer. This is because the order cost is lower (originally 3.70 for
ordering 100 or more at a time but now 2.55 for ordering 175 at a time) so she
cannot be absolutely sure that the new offer is more costly without doing a
numeric calculation.

As the two alternatives here specify precisely the order quantity we need
only calculate the cost for these two alternatives.

Ordering Q=175 we have the cost per year is RP + (chQ/2) + (coR/Q)

=1560×6 + (0.42×175/2) + (2.55×1560/175)
= 9419.5

Ordering Q=750 we have the cost per year is RP + (chQ/2) + (coR/Q)

=1560×5.5 + (0.07×5.5×750/2) + (2.55×1560/750)
= 8729.7

From part (b) of the question ordering 166 at a time had a cost of 9429.6.
Hence ordering 750 at a time is the cheapest option over all order quantities
costed above.

Question 6
An organisation has six branches (A to F) that have been operating for many
years. As a result of the drop in sales over the COVID pandemic it has decided

13
MN2032 Management science methods

that two of these six branches should be closed. Data (given in the table below)
has been collected for these branches with regard to the number of customers
per employee (in a past year) and sales per employee (in a past year). Here the
past year is the financial year April 2021 to March 2022, as audited figures for
the year commencing April 2022 are not yet available.

Customers (‘000) per Sales (£’0000) per

Branch employee employee

A 3.4 1.3
B 6 1.3
C 2.3 0.6
D 3.2 0.9
E 0.9 0.5
F 0.8 1

For example, branch E last year had 900 customers per employee and sales per
employee of £5000.
a) Considering the data given above which two branches do you suggest be
closed? Give your reasons.

b) Apply data envelopment analysis to compare the relative performance of

these branches using the data shown above. Copy the table shown below
and fill in the efficiencies and reference sets (where appropriate). Which
two branches do you suggest be closed? Give your reasons.

Branch Efficiency (% integer) Reference set

A
B
C
D
E
F

c) A colleague has suggested that the data used to decide the branches to
close will have been unduly influenced by COVID, the impact of which
branches could not have anticipated. How would you respond to this and
how might this affect your decision in part (b) as to which two branches to
close?

Reading for this question

This question relates to Chapter 9 of the subject guide.
Approaching the question

14
 Examiners’ commentaries 2023

a) The two branches with the worse performance for either ratio are F (first ratio)
and E (second ratio), so recommend these should be closed.

b) The ratios given in the question can be regarded as Output/Input, so are

appropriate to use. Then with respect to these ratios the relative efficiencies
are:

Branch Customers per employee Sales per employee

A 0.57 1.00
B 1.00 1.00
C 0.38 0.46
D 0.53 0.69
E 0.15 0.38
F 0.13 0.77

So here, A and B are the best performers with respect to the second ratio, B is
the best performer with respect to the first ratio. The DEA diagram is:

The required table is:

Branch Efficiency Acceptable range Reference

(% integer) set
lower upper

A 100
B 100
C 46 44 48 A,B

15
MN2032 Management science methods

D 69 67 71 A,B
E 38 36 40 A
F 77 75 79 A
The acceptable range above is the range within which the examiners count an
answer as correct.

So now recommend the two branches with the lowest efficiencies for closure,
so here C and E

c) Here the issue is how COVID may have impacted the last audited year figures
and how that affects any closure decisions. Points that the examiners would
expect to see here include:

• wait for audited figures for the financial year from April 2022
• if we assume that all branches were equally affected by COVID this makes
no difference to any closure decisions
• we could investigate how the branches chosen for closure were affected by
COVID to see if they were more badly affected by COVID than other
branches.

Question 7
a) The payoff table below shows for a company the final year end profits
expected (£m) from four possible choices (A, B, C and D) with respect to
three possible scenarios (S1 to S3 respectively) for the forthcoming year.

S1 S2 S3
A -1 1 3
B 16 -3 18
C 1 14 18
D 2 10 -5

Here, for example, if the company makes choice B (and it can only choose
one of A, B, C or D) and the scenario for the forthcoming year turns out to
be S3 then it makes a profit of £18m. The company estimates that the
probabilities associated with scenarios S1, S2 and S3 occurring are 0.20,
0.35 and 0.45 respectively.

Copy the following table and fill in the choice that should be made under
each criterion and the associated value.

Associated value £’m

Criteria Choice (2dp)

Optimistic

Conservative

16
 Examiners’ commentaries 2023

Regret

Equally likely

EMV

Based on your analysis which choice from A, B, C or D would you

recommend the company makes? Clearly state the single major reason for
the choice that you are recommending.

b) A colleague has argued that your analysis is irrelevant as any sensible

person knows that the general rule to be applied in all situations is that if
there is a unique choice that gives a positive profit for all scenarios you
choose it. Using this rule choice C is the obvious choice. As such choices
A, B and D can be instantly discarded. How would you respond to this
argument?

c) In a single server queueing system with an exponential service time

distribution and a Poisson arrival distribution customers arrive at an
(average) rate of 80 per hour. The server takes (on average) 40 seconds to
serve a customer.

Copy the following table into your answer and fill in the corresponding
numerical values.

Factor Value (4dp)

Traffic intensity
Average number of customers in the
queue

Probability that there are three or more

customers in the system

Reading for this question

Parts (a) and (b) relate to Chapter 4 of the subject guide and part (c) relates to
Chapter 11 of the subject guide.
Approaching the question
a) The situation is:

Probabilities 0.2 0.35 0.45

Scenario
Payoff Choice S1 S2 S3 Equally likely EMV
A -1 1 3 1.00 1.50
B 16 -3 18 10.33 10.25
C 1 14 18 11.00 13.20
D 2 10 -5 2.33 1.65

Regret
Choice S1 S2 S3 max regret
A 17 13 15 17.00
B 0 17 0 17.00

17
MN2032 Management science methods

C 15 0 0 15.00
D 14 4 23 23.00

The required table is:

Criteria Choice Associated value 2dp

Optimistic B,C 18
Conservative C 1
Regret C 15
Equally likely C 11.00
EMV C 13.20

So here C the recommended choice as it the majority choice (taking joint

choices into account).

b) This general rule is obviously incorrect. Note that the rule is ‘if there is a unique
choice that gives a positive profit for all scenarios you choose it’.

For example in the current payoff matrix if scenario S1 had probability 1 and so
the other scenarios will never occur (meaning we are restricted to the first
column in the payoff matrix) we will choose B over C.

The fact the rule stated in the question chooses C is irrelevant, it is just a wrong
rule.

c) Here we have:
λ=80 per hour
µ=3600/40=90 per hour

Using these values and the formulae in the subject guide (page 200) the
required table is:

Factor Value (4dp)

Traffic intensity 0.8889
Average number of customers in the queue 7.1111
Probability that there are three or more customers in 0.7023
the system

Here the last value in this table calculated by finding 1- prob(exactly 0 in

system) –prob(exactly 1 in system) - prob (exactly 2 in system).

Question 8
Briefly explain each of the following topics from the viewpoint of Operational
Research/Management Science:
a) MIP
b) Value judgements in DEA
c) Project time/project cost trade-off in network analysis

Your answer to this question should be limited to 2 pages for each of the three
parts (so at most 6 pages in total).

18
 Examiners’ commentaries 2023

Reading for this question

Part (a) relates to Chapter 7 of the subject guide; part (b) relates to Chapter 9 of the
subject guide and part (c) relates to Chapter 3 of the subject guide.
Approaching the question
a) Points that the examiners would expect to see here include:
• MIP stands for Mixed Integer Programming
• Some variables take integer values
• Some variables take continuous (fractional) values
• Single objective
• Objective optimised subject to constraints
• Constraints/objective linear

b) Points that the examiners would expect to see here include:

Value judgements in DEA is to do with:

• adding additional constraints (typically to the weights)
• done so as to better represent the situation being modelled.
• may introduce issues relating to solving nonlinear problems

c) Points that the examiners would expect to see here include:

Project time/project cost tradeoff in network analysis) relates to:

• crashing (time/cost tradeoff)
• normal point and crash point
• linear relationship between normal point and crash point (or diagram)
• use of linear programming
• LP objective is minimise cost subject to achieving desired project completion
time.

19