MSIN0159 Decision and Risk Analysis Examination Paper 2021/22
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
MSIN0159 Decision and Risk Analysis
Examination Paper
2021/22
SECTION A
This section consists of THREE (3) questions and is worth THIRTY (30) marks.
XMicron,a high-end semiconductor manufacturer, spent £5 million to purchase a single machine to produce chips. The machine produces 800 chips per hour. Upon production, each chip is immediately inspected for defects. Xmicron knows that the chance of each chip being defective is independent and identical, and estimates this chance as being 2.5%. The defective chips are discarded and the rate of production is constant over time.
The machine needs to fulfil all chip orders that come in for that day. The average number of orders per day is 7800 chips. However, daily orders received is highly uncertain. Specifically, historical data shows that the number of orders per day has a triangular distribution, with minimum, most likely and maximum values of 5000, 8400 and 10000 respectively. The orders cannot be fulfilled by using defective chips.
According to the production manager, David Dunn, the maximum number of hours the machine can work in a day is 12 hours. Then, the machine needs to be turned off until the next day to cool down.
David Dunn argues that since the average number of working chips produced is 780 per hour, it will take the machine on average 10 hours in a day to fulfil the orders. This number is considerably lower than maximum number of hours the machine can work in a day (i.e. 12 hours). Therefore, he believes that the machine will be able to meet demand on every single day.
Question A.1:
What are the possible probability distributions that can be used to model the number of defects produced by the machine per hour? Justify your answer and list all parameters of the distributions you specify. [10 marks]
Question A.2:
What do you think of David Dunn’s reasoning? Is the average time the machine needs to run truly 10 hours? Will it be able to meet demand on every single day? [10 marks]
Question A.3:
How would you build a simulation model to estimate the average number of days per year that the machine would not be able to fulfil all customer orders within 12 hours? While describing your model, suppose that you use SimVoi to create your simulation model in Excel. (You should not actually create your model in Excel. Just describe how you would build it in Excel by using SimVoi). [10 marks]
SECTION B
This section consists of THREE (3) questions and is worth THIRTY (30) marks.
Armour, a petroleum company, is in the process of diversifying their business in preparation for the global shift away from fossil fuels. Until they can switch away wholly from fossil fuel production, they have made it a strategic priority to create less polluting gasoline blends. Their R&D team has just come up with a new petroleum product, codenamed ChemicalX, that has an octane rating of 120, and would emit 100 pounds of greenhouse gases per million BTU of energy produced. ChemicalX costs £1 per litre to produce. They intend to create a blend of ChemicalX and their standard petroleum product. The standard product has an octane rating of 100, emits 200 pounds of greenhouse gases per million BTU of energy produced, and costs £0.75 per litre to produce.
Armour wants to create a petroleum blend (a blend of ChemicalX and the Standard product) that has an octane rating of at least 112, and that emits at most 145 pounds of greenhouse gases per million BTU of energy produced. They have enlisted your help to figure out how best to create this blend. What ratio (by volume) should the two products be mixed in?
Note: You may assume that mixing the two products results in an octane rating that is the weighted average of the individual octane ratings, with the weights being the volumes. Similarly, you may assume that the emissions of the resulting product is the weighted average of the emissions of the individual products, with the weights being the volumes.
Question B.1:
Formulate a linear optimisation model that can be used to solve Armour’s decision problem. Formulate the problem with as few decision variables as possible. Describe briefly what each of your decision variables stands for. In terms of the decision variables, list the objective and constraints formally. [10 marks]
Question B.2:
Inspect the optimisation problem you formulated in Question B.1. What is the optimal ratio to mix the two products in? Explain intuitively how you obtained this solution. (You should not and do not need to use Solver to answer this question. If you have time you may, of course, use Solver to confirm your answer). [8 marks]
Question B.3:
Figure 1 shows a snippet of the sensitivity report for the above optimisation problem. On the basis of the sensitivity report, answer the following questions.
Figure 1: Sensitivity Report
a) Which of the constraints is binding in the optimal solution? [2 marks]
b) If the octane requirement is increased to 115, how much would the objective change? Justify your answer. [4 marks]
c) Explain intuitively why the emissions requirement constraint has such a large allowable increase. [6 marks]
SECTION C
This section consists of FOUR (4) questions and is worth FORTY (40) marks.
The EnergyX Oil Company currently has an option to purchase a piece of land with an oil well on it. It is now May 1st, and the current price of the land is $2.4 million. EnergyX will not be able to make use of the oil well from this land until the beginning of July, however, it is worried that that another company might purchase the land before the beginning of July. It believes that there is a 25% chance that a competitor will buy the land during May. If this does not occur, EnergyX estimates that there is a 20% chance that a competitor will buy the land during June. If EnergyX does not purchase the land now, it can attempt to purchase it at the beginning of June or the beginning of July, provided that the land is still available.
If EnergyX delays the purchase of the land, the price of the land may increase or decrease in the next two months. In May, the price will increase by $60,000 with probability 0.6, and decrease by $120,000 with probability 0.4. In addition,
if the price increases by $60,000 in May, then the possible
subsequent price decreases in June are $50,000, and $60,000 with respective probabilities 0.4 and 0.6;
and if the price decreases by $120,000 in May, then the possible
subsequent price decreases in June are $20,000, and $100,000 with respective probabilities 0.7 and 0.3.
If EnergyX purchases the land at any time, it believes that it can gross $3.2 million (excluding the cost of purchasing the land). But if it does not purchase the land, EnergyX believes that it can make $650,000 from an alternative investment. This alternative investment opportunity is available at any time.
Question C.1:
Construct the decision tree for EnergyX’s decision problem. Assuming it is risk-neutral, what would you recommend EnergyX to do? How much expected profit does EnergyX make under your recommendation? [10 marks]
Question C.2:
a) Define the risk profiles associated with purchasing the land in May and delaying the purchase till June. Taking risk into consideration and assuming EnergyX is risk-averse, what would be your recommendation? How would your answer change if EnergyX is risk-loving? [4 marks]
b) EnergyX is not sure about its current estimates for the probability of competitor purchasing in May and the probability that the land price increases by $60,000 in May. EnergyX wants to check the robustness of the optimal strategy obtained in Question C.1 with respect to these two probabilities. Figures 2, 3 and 4 below show the results of one-way and two-way sensitivity analyses. What can you conclude from these three figures? [6 marks]
Figure 2: One-Way Sensitivity Analysis of the Profit from Purchasing the Land in May and Waiting till June with Respect to the Probability that the Competitor Purchases the Land in May
Figure 3: One-Way Sensitivity Analysis of the Profit from Purchasing the Land in May and Waiting till June with Respect to the Probability of $60,000 Price Increase in May
0 0.025 0.05 0.075 0.1 0.125 0.15 0.175 0.2 0.225 0.25 0.275 0.3 0.325 0.35 0.375 0.4 0.425 0.45 0.475 0.5 0.525 0.55 0.575 0.6 0.625 0.65 0.675 0.7 0.725 0.75 0.775 0.8 0.825 0.85 0.875 0.9 0.925 0.95 0.975 1
1
0.975
0.95
0.925
0.9
0.875
0.85
0.825
0.8
0.775
0.75
0.725
0.7
0.675
0.65
0.625
0.6
0.575
0.55
0.525
0.5
0.475
0.45
0.425
0.4
0.375
0.35
0.325
0.3
0.275
0.25
0.225
0.2
0.175
0.15
0.125
0.1
0.075
0.05
0.025
0
0 0.025 0.05 0.075 0.1 0.125 0.15 0.175 0.2 0.225 0.25 0.275 0.3 0.325 0.35 0.375 0.4 0.425 0.45 0.475 0.5 0.525 0.55 0.575 0.6 0.625 0.65 0.675 0.7 0.725 0.75 0.775 0.8 0.825 0.85 0.875 0.9 0.925 0.95 0.975 1
Probability of $60,000 Price Increase in May
Figure 4: Two-Way Sensitivity Analysis of Purchasing the Land in May and Waiting till June Strategies with Respect to the Probability that the Competitor Purchases the Land in May and the Probability of $60,000 Price Increase in May
Question C.3:
a) Suppose that EnergyX has insider information about its competitor and can learn before May 1st whether the competitor will purchase the land in May or not. How much is such information worth? Why? [6 marks]
b) Suppose that EnergyX does not have insider information about its competitor (i.e. ignore part a) but that it can block its competitor from purchasing the land in May. What is the maximum amount EnergyX would be willing to pay to control whether its competitor purchases the land in May or not? How is that value compared to the value of information you found in part a? Why? [4 marks]
Question C.4:
Suppose that, in addition to having perfect information about whether its competitor will purchase the land in May or not as in Question C.3a, EnergyX can hire a data analytics firm which can perfectly predict whether the land price will increase by $60,000 or decrease by $120,000 in May. Determine the value of additional (perfect) information on the change of land’s price in May assuming EnergyX is risk-neutral. [10 marks]
2023-03-22