Decision Support Systems

Midterm Examination

Spring 24

INSTRUCTIONS:
1) You must answer the critical reflection question (10 points).
2) Answer two other questions from Part A (10 points each).
3) Solve three questions from Part B (application-oriented questions, 20 points each).
4) There are 10 additional points for quality of presentation.

The completed exam must be uploaded no later than Sunday, February 25 at 11:55PM.

Completed exams must be uploaded to the assignment section preferably in a zip file labeled
last_first_midtermexam.
Your submission includes:
- This file with your answers to the chosen essay questions. All essays need references.
Sign the Academic Honesty Statement (page 2)
- All Excel files with written out final answers.

In preparing your answers, you may use and reference course material, published sources, the
World Wide Web, and/or your own notes. However, you may not ask for or receive help from
any person on the exam. The use of ChatGPT or other AI tools must be indicated.

A comment on academic honesty (may sound obsessive, but it is good advice):

The nature of online work sometimes leads to questionable decision making. This is an exam,
and therefore it is individual work. Please avoid the temptation of working in groups. It is not
only morally wrong, but also myopic, as group work presented as individual work stands out, no
matter how much you try to disguise it.

Please, don’t risk it. It is not worth it. If I suspect a case of academic dishonesty, all
involved parties will lose points up to complete failure (grade F) of the exam.

Best of success!!

Page 1
DECISION SUPPORT SYSTEMS MIDTERM EXAM Spring

Decision Support Systems

Midterm Examination – Spring 24

I hereby certify that I have completed the attached examination materials, using only my own
efforts. I have not asked for or received help from any person in completing this exam.

________Luca Bulfon_________ Feb. 25th 2024

(Name) (date)

Page 2
DECISION SUPPORT SYSTEMS MIDTERM EXAM Spring

Part A – Mini-Essay oriented questions. Write your answers under the question you choose
and delete the ones not answered.

1) Critical reflection question (required): A term showing up a lot in current discussion is

that we live in a “VUCA” world (volatile, uncertain, complex, ambiguous). Based on
academic sources, discuss if traditional decision-making approaches fit to this decision-
making environment (not more than 500 words).

Although the concept of VUCA originated in the military many years ago, it is now
becoming a very diffused topic of discussion, especially in the business world, due to two
major factors: globalization and the technological advancement. These factors, combined
with the speed of innovation and the increasing competition are causing the world to
experience quick changes and turns in all sectors of the economy. The original concept of
VUCA was created to describe the inability of Cold War countries to adapt to the new
and fragmented organization of the post World War world. The same could be said today
about the traditional decision-making processes compared to the new agile methods: the
recent pandemic was an example on how even some big companies were too slow to
react to sudden changes and had to pay the price for it. This doesn’t mean that greater
VUCA is the origin of problems, it should instead be seen as a source of opportunities, in
which quick and good decision making can beat the competition and grow. I have
recently listened to an interview of Jeff Bezos which relates to decision making in a
VUCA world: he was explaining how he used to make decisions when he was CEO of
Amazon. He mentioned that “Most decisions should probably be made with somewhere
around 70% of the information you wish you had. If you wait for 90%, in most cases,
you're probably being slow.” As changes could happen so quickly in today’s world,
counting on traditional decision-making models like the ones we are studying in class can
be very dangerous, as they don’t take into consideration external factors which can
completely change any decision making scenario.

Larson, Erik. "Is VUCA Really The Enemy Of Good Decision-Making?" Forbes,
28 Dec. 2021, www.forbes.com/sites/eriklarson/2021/12/28/is-vuca-really-the-enemy-of-
good-decision-making/?sh=6c080ad054ea.

Indigo Anchor. "On the Origins of VUCA and How it Affects Decision Making."
Indigo Anchor, 31 Oct. 2019, www.indigoanchor.com/blog/2019/10/31/on-the-origins-
of-vuca-and-how-it-affects-decision-making.

2) Describe the anchoring and framing biases and give examples of each of them that are
not identical to those from our course material.

Both anchoring and framing bias are cognitive tricks that can affect our decision making.
The anchoring trap happens when our mind gives too much weight on information about
past events, trends, comments, statistics, or information carried within the situation itself.
An example of can be shopping online, especially on Amazon and we see products
advertised as great offers. For example, I was looking for an electric shaver that was
showing on sale from $89.99 to $69.99. Then I downloaded a price tracker plug in that

Page 3
DECISION SUPPORT SYSTEMS MIDTERM EXAM Spring

showed me the price at which this item was sold, and from the moment it was put on
Amazon it has always been sold at around $75. In this case, seeing the price of $89.99
acted as an anchor, influencing my perception of possibly making a “good deal”.
The framing bias refers to how the presentation or the wording of information can
influence our decision-making. We are also constantly exposed to this type of bias from
the media and in the newspaper. For example, the news of the day could be that
“Unemployment is at 5%”, but two different news report one that "Unemployment Rate
Falls to 5%.", and another "Job Losses Continue as Unemployment Rate Reaches 5%.".
Readers exposed to different headlines might develop contrasting opinions about the state
of the economy based solely on the way the information is framed, and this could be very
powerful, especially in periods close to the elections.

3) Below is the output of a sensitivity report for an LP problem. The objective of the
problem is to find the combination of products (A/Cs and fans) that maximizes profit.
Constraints are the available wiring and drilling hours, as well as a minimum number of
A/Cs produced, and a maximum of fans. Explain what the sensitivity report tells you
about the model. Specifically, answer the following questions:
a. Explain the meaning of the reduced cost of zero for both products.
b. What does the allowable increase/decrease for the number of units of A/Cs mean?
c. What does the shadow price of five mean for wiring hour constraint?
d. What does the shadow price of zero for the constraint that requires a minimum of
A/C production mean?

Objective Cell (Max)

Final
Cell Name Value
$D$6 Profit 1900

Decision Variable Cells

Final Reduced Objective Allowable Allowable
Cell Name Value Cost Coefficient Increase Decrease
Number of Units
$B$5 A/C 40.0 0.0 25 5.000000033 2.500000025
Number of Units
$C$5 Fans 60.0 0.0 15 1.666666683 2.500000017

Constraints
Final Shadow Constraint Allowable Allowable
Cell Name Value Price R.H. Side Increase Decrease
$D$8 Wiring Hours 240.0 5.0 240 10 30
$D$9 Drilling Hours 140.0 5.0 140 20 6.666666667
$D$10 Minimum A/C 40.0 0.0 20 20 1E+30
$D$11 Maximum Fans 60.0 0.0 80 1E+30 20

a. The reduced cost indicates the change in the objective function's value per unit increase in
the corresponding variable. A reduced cost of 0 means that increasing the any of the two
variables by one unit (in this example producing one more A/C or fan), would not change
the objective function, therefore it would not make any sense.

Page 4
DECISION SUPPORT SYSTEMS MIDTERM EXAM Spring

b. Allowable Increase/Decrease indicates the range of change in unit profit for ACs or fans
that would not change the optimal solution. For example, if the profit per unit of A/Cs
increases by no more than $5 or decreases by no more than $2.50, it would still be
optimal to produce 40 units. The same is applied to fans ($1.66<= change<= $2.50).
c. The shadow price is the marginal value of having an extra unit of a particular resource. In
this case, for wiring hours, all the resources available are being used, but if I could add
one extra hour to the right-hand side of this constraint, I would increase the profit by 5
dollars. Of course, this is only valid if it is within the Allowable Increase range.
d. A shadow price of zero means that the constraint is nonbinding and that increasing the
right-hand side of Maximum Fans would not provide any increase in profit. This is why it
has an infinite Allowable Increase and an Allowable Decrease of 20 which is the
difference between 80 and 60.

Page 5
DECISION SUPPORT SYSTEMS MIDTERM EXAM Spring

Part B – Application oriented questions. Build the models and answer the questions within
the Excel files.

1. Given the following linear programming setting, approximately reverse-engineer the

Objective function and the constraints from the graph; solve it graphically, finding an
estimate for X1 and X2. Assume that the objective function seeks maximization.

2. Bob is preparing the sales forecast for his newly launched company over 6 periods. He has
the following data:
- Sales for period 1 = $100,000; estimated growth rate per period, uniformly distributed
between 5% and 8%.
- Cost of sales for period 1 = $80,000; estimated growth rate per period, normally
distributed with a mean of 8% and std deviation 3%.
Using simulation (1000 trials) estimate the average total gross profit after 6 periods, and its
standard deviation (Note: total gross profit = total sales – total cost of sales).

Page 6
DECISION SUPPORT SYSTEMS MIDTERM EXAM Spring

3. A car dealership needs to determine how to allocate its $20,000 advertising budget. They
have estimated the expected profit from each dollar (X) spent in four different advertising
media as follows:

Medium Expected Profit

Newspaper 100 * X0.7
Radio 125 * X0.65
TV 180 * X0.6
Direct Mail 250 * X0.5

If the company wants to spend at least $500 on each medium, how should it allocate its
advertising budget to maximize profit?

4. “Paints for the planet” produces 2 types of paints, outdoor and indoor. The outdoor line
sells for $40 a gallon, with production costs of $25, the indoor line for $28, with
production costs of $15. Each gallon of paint goes through two processing steps: pigment
mixing and final paint mix. The table below summarizes the hours needed. There are
2,400 hours of pigment mixing time available, and 1,500 hours of final paint mix. The
company wants to maximize profit, while also minimizing waste water usage. A gallon of
outdoor paint requires 45 gallons of water during production, while indoor requires 30.
The company set itself a limit of not producing more than 15,000 gallons of waste water
per production run. Furthermore, the marketing team has decided that demand for the
outdoor line is at least 100 gallons. How many of each type of paint should “Paints for
the planet” produce in order to maximize profit and keep its environmental promise?
Formulate the ILP model, implement it in Excel and solve it. Discuss the results for the
constraints (you do not have to create a sensitivity report to do so).

Process step Outdoor Indoor

Pigment mixing 2 3
Final Paint mix 4 1

Page 7