MAX 502 Homework 10 Prof.
Panchal
MBA 504 Fall 2023
Problem 1: Revisit the dice problem you modeled with @Risk in the last problem set.
• Run the simulation. Make sure that the iterations are set to 1,000, and the simulations are set
to 1.
• Use RiskTarget and RiskTargetD to find the probabilities of a Win, Loss and Tie.
2
Problem 2: This problem is similar to the Inventory Management exercise we did in class.
Andrew Molina just bought a new hotdog stand at the local ballpark.
• Each hotdog will sell for $5.95
• It costs Franks $3.95 to make each hotdog (served with fancy napkins).
• Hotdogs not sold by the end of each game are reduced to $3.50 and sold on-clearance at the
train station located outside of the stadium.
Andrew believes that game-demand for hotdogs follows a triangular distribution with the most likely
demand being 8,500; the minimum demand being 7,500 and the maximum demand being 10,0001. Fans
love hot dogs in Chicago!!
a. Suppose Andrew decides to make 8,750 hotdogs. How much profit can Frank expect, on
average?
• To answer this question, simulate Franks’ profits over 1,000 games (iterations).
• To generate the game demand for hot dogs, use the RiskTriang(min, m.likely, max)
function, where min refers to the minimum observable value, and max refers to the
maximum observable value, and m.likely refers to most likely demand value.
• Calculate the profit earned in a game, based on Franks decision to stock 8,750 hotdogs for
each game, and the realized demand for the game.
• Calculate the expected (or mean) profit across the 1,000 simulated games. Use riskmean.
b. Replicate the model you created in part a. to calculate the expected profit for stocking levels
between 7,500 and 10,000, in increments of 250.
What should Andrew stock to maximize his profits?
1RiskTriang(minimum,m.likely,maximum) can be skewed in either direction, depending on the position of m.likely relative to
minimum and maximum. This distribution is perhaps the most readily understandable distribution for basic risk models. Its
possible drawback for real applications is that it allows no possible values outside the min-max range.
3
Problem 3: Recall the Teal-Star-Sapphire Moon of Illinois problem of Problem Set #10.
Teal-Star-Sapphire-Moon of Illinois is an insurance company that offers health insurance plans to the employees of
Illinois-based companies. Employees of subscriber firms have a choice among three different plans, each with
various degrees of coverage, from a low-cost plan with a high deductible to a high-cost plan with no deductible.
The details for each plan are given in the following table.
Plan Options and Costs
Maximum
DEDUCTIBLE Out-of-Pocket Co-Insurance PREMIUM
(Annual Amount) (after Deductible) (after Deductible) (Cost per Month)
2-person 3+person
1-Person Family 1-Person Family 1-Person Family Family
Bronze $7,400 $17,100 $7,400 $17,100 0% $315 $509 $702
Silver $2,200 $5,000 $8,550 $10,600 50% $483 $711 $784
Gold $750 $3,750 $8,550 $12,600 25% $687 $728 $972
The annual premium is the amount the employee pays to buy the insurance plan. In the table above, premium
amounts are shown as a monthly cost. The annual deductible is the amount the employee pays for covered health
care services before his/her insurance plan starts to pay for medical expenses. The co-Insurance (or co-payment
percentage) is the percent of the medical costs the employee must cover once the deductible has been paid. The
maximum out-of-pocket cost details the maximum amount the individual will be responsible for in medical costs
for the year, after paying the deductible and premium.
Based on the application developed in the Problem 3 worksheet, use @Risk to model the projected medical costs
under each of the possible plans for an employee (signing up for a Family Plan) whose expected household medical
expenses follow a Normal distribution with a mean of $20,000 and a standard deviation of $6,000.
Consider an employee (signing up for a Family Plan) whose expected household medical expenses follow a Normal
distribution with a mean of $20,000 and a standard deviation of $6,000.
4
Hints:
• Set the Projected Expenses as an Input. Use the Normal Distribution. It can be found under the
continuous Distributions. Set the mean = 20000 and standard deviation= 6000.
• Set each of the plans’ totals as output cells. Select all three cells (B19:D19), and click Output.
• Simulate Profits using 1,000 iterations.
a. Which plan has the lowest expected cost? What is the expected cost?
a. Under each cost, place the mean expected cost. You will be able to see this by going to the
Explore button, and clicking on Browse results.
•
• To write these values in the workbook,use the RiskMean function.
o You can find it under the Define Group, under the Fx button, in the Simulation Results
row.
• Write the expected (or mean) cost for each plan in cells B21:D21. You must refer to each cost cell
when using the RiskMean function. That is, to get the expected cost for the Bronze Plan, we
would type: =RiskMean(B19).
5
b. Use Statistics Functions of @Risk to show the minimum and maximum costs associated with each plan.
That is, use the RiskMin and RiskMax Functions. Place the results in B22:D23.
c. For each plan, enter the probability that the total cost associated with the plan will be more than $18,000.
• You can do this by just moving the arrows at the top of the output charts, and writing the
answers in the workbook or by using the RiskTargetD function. It can be found using the Insert
Functionsè Simulation Resultsè RiskTargetD. Look at the help menu to understand what this
function does.
6
Problem #4: Spencer Airways offers commuter flights between small cities in upstate New York. Once
per day, they offer a flight from Rochester to Buffalo using a six-passenger commuter plane. The one-
way fare for this flight is $129 and is non-refundable if the person does not show up for the flight.
The daily demand for this flight is given below, along with the probability distribution for no-shows (that
is, the number of passengers who will not show up for the flight):
Demand Probability No Shows Probability
5 5% 0 15%
6 11% 1 25%
7 20% 2 26%
8 18% 3 23%
9 16% 4 11%
10 12%
11 10%
12 8%
To compensate for no-shows, Spencer’s policy is to overbook (that is, accept reservations from extra
passengers) beyond the plane’s capacity of 6. If it turns out that there are not enough seats to
accommodate everyone, Spencer refunds those individuals’ fares and provides each person with a $150
voucher good on any future flight. Currently, the firm’s policy is to overbook three passengers per flight
whenever possible. Thus, if Spencer Airways sells 8 tickets for a given day, and there is 1 no show for
the flight, the airline will pay $279 to one (8 – 1 – 6) passenger, and ask him/her to fly on a different day.
The fixed cost per flight is $450, regardless of the number of passengers.
a. Set up a base model where, based on the overbooking policy and plane capacity, the inputs are
the demand for tickets for the flight and the number of no-shows. Calculate the profit per flight.
b. Next, using the model from part a., create a simulation model that will estimate Spencer’s
expected profit per flight. Given the costs, distributions listed above, what expected profit can
the firm expect?
c. Currently Spencer’s policy is to overbook three passengers per flight. The firm is wondering
whether this number is optimal. Using the RiskSimTable function, investigate the possibility of
overbooking 0, 1, 2, 3, 4 and 5 passengers. What is your recommendation and why?
