$3.$ CRYSTAL VISION $(25$ POINTS$)$

Crystal Visions $($CV$)$ has designed a new model of video projector$$the Home Theater $5000($HT$5000).$ A key

component of the HT$5000$ is its high$-$resolution LCD panel. CV is considering whether to buy the LCD panels from

an outside supplier or set up and manufacture the LCD panels themselves. CV can purchase the LCD panels for

$$200$ each from an outside supplier. If they instead make the LCD panels themselves, there would be a $$1\mathrm{.}3$ million

cost to set up the manufacturing facility, but the unit cost would then be just $$160$ each. If demand for the HT$5000$ is

high, CV projects they will need $50,000$ LCD panels. If demand for the HT$5000$ is low, CV projects they will only

need $20,000$ LCD panels. CV believes there is a $50\%$ chance that demand will be high, and a $50\%$ chance demand

will be low. However, they won$$t find out the level of demand until well after the HT$5000$ is in production and thus

well after they have committed to whether they will be making or buying the LCD panels.

Suppose that before deciding whether to make or buy the LCD panels, CV has the option to conduct a survey at a

cost of $$100,000$ to assess consumer attitudes about the HT$5000.$ The results of this survey would indicate either a

favorable or unfavorable attitude toward the HT$5000.$ CV has conducted such a survey for other similar products

forty times in the past. Typically, $($but not always$),$ a favorable attitude on the survey has meant demand will be high,

while an unfavorable attitude has typically $($but not always$)$ meant demand will be low. The results for the past forty

times the survey have been used are shown below.

Requirements of decision tree model format $($follow the instructions below to avoid losing credits$)$:

$$ Show all given parameters $($unit cost$,$ setup cost, market probabilities, etc.$)$ in data cells next to the decision tree

and always refer to those data cells wherever they are used in the decision tree.

$$ Show all relevant probabilities $($e$.$g$.,$ cond