ISYE 6644 Week 7 Homework
Attempt History
Score for this quiz: 15 out of 15
Submitted Oct 10 at 8:40pm
This attempt took 1 minute.
Question 1
pts
(Lesson 5.15: Two-Channel Manufacturing Example.) Consider the demo from class:
Module05-15 - Model 04-01-ElectronicAssembly-slow.doe
If you run the thing for a while, you’ll see that the rework Station gets pretty clogged up every once
in a while – probably because the service times are EXPO(45) minutes. Let’s suppose that
management has noticed this as well and decides to upgrade that station so that the service times
are now EXPO(10). Go ahead and make this change and comment on what you see.
a. There are still very long lines at Rework almost all of the time.
b. There are occasionally extremely long lines at Rework.
c. There are occasionally short lines, but long lines are very rare.
d. There is never, ever any line at Rework (c).
Question 2
pts
(Lesson 5.15: Two-Channel Manufacturing Example.) To get you in the mood for this problem, let’s
recall the slightly different demo that incorporates a logic-based equation in the Sealer Process
module:
Module05-15 - Model 04-01-ElectronicAssembly-Logic-slow.doe
I’m not going to ask you to run this model, but I’d simply like to know the value of the related logic
expression,
( (1 == 1) * 10 ) + ( (1 == 0) * 5 )
a. 0
b. 1
c. 5
, d. 10
e. 15
Since “1 == 1” is true and “1 == 0” is false, we have (
(1 == 1) * 10 ) + ( (1 == 0) * 5 ) = (1*10) + (0*5) =
10, so that the answer is (d).
Question 3
pts
(Lesson 5.16: Fake Customers.) Let’s use fake customers to calculate an exponential probability. In
particular, modify the following demo from class so that you can estimate the probability that an
EXPO(2) random variable (this is the Arena notation for an exponential RV with mean 2) is > 2.
Module05-16 - Model 04-05 – NormalProb.doe
Hints on how to do the necessary modification:
Go to the Assign module and change the normal RV to the exponential.
Optional: You might also want to change the name of the fake customer’s attribute from “Normal
Observation” to “Exponential Observation” just to make things clearer.
Go to the Decide module and make the obvious change to the condition.
Optional: You might also want to change the counter names in the Record modules to something
more obvious.
Let the run complete and then look at the output! Note that the demo runs for 1 million samples,
so the estimate ought to be very good.
a. 0
b. about 0.135
c. about 0.368
d. about 0.632
e. about 0.865
Solution: The answer is (c). In fact, when I ran my simulation, 367460 out of the 1 million fake
customers satisfied the inequality.
Of course, you can calculate the answer exactly (without using simulation):
Attempt History
Score for this quiz: 15 out of 15
Submitted Oct 10 at 8:40pm
This attempt took 1 minute.
Question 1
pts
(Lesson 5.15: Two-Channel Manufacturing Example.) Consider the demo from class:
Module05-15 - Model 04-01-ElectronicAssembly-slow.doe
If you run the thing for a while, you’ll see that the rework Station gets pretty clogged up every once
in a while – probably because the service times are EXPO(45) minutes. Let’s suppose that
management has noticed this as well and decides to upgrade that station so that the service times
are now EXPO(10). Go ahead and make this change and comment on what you see.
a. There are still very long lines at Rework almost all of the time.
b. There are occasionally extremely long lines at Rework.
c. There are occasionally short lines, but long lines are very rare.
d. There is never, ever any line at Rework (c).
Question 2
pts
(Lesson 5.15: Two-Channel Manufacturing Example.) To get you in the mood for this problem, let’s
recall the slightly different demo that incorporates a logic-based equation in the Sealer Process
module:
Module05-15 - Model 04-01-ElectronicAssembly-Logic-slow.doe
I’m not going to ask you to run this model, but I’d simply like to know the value of the related logic
expression,
( (1 == 1) * 10 ) + ( (1 == 0) * 5 )
a. 0
b. 1
c. 5
, d. 10
e. 15
Since “1 == 1” is true and “1 == 0” is false, we have (
(1 == 1) * 10 ) + ( (1 == 0) * 5 ) = (1*10) + (0*5) =
10, so that the answer is (d).
Question 3
pts
(Lesson 5.16: Fake Customers.) Let’s use fake customers to calculate an exponential probability. In
particular, modify the following demo from class so that you can estimate the probability that an
EXPO(2) random variable (this is the Arena notation for an exponential RV with mean 2) is > 2.
Module05-16 - Model 04-05 – NormalProb.doe
Hints on how to do the necessary modification:
Go to the Assign module and change the normal RV to the exponential.
Optional: You might also want to change the name of the fake customer’s attribute from “Normal
Observation” to “Exponential Observation” just to make things clearer.
Go to the Decide module and make the obvious change to the condition.
Optional: You might also want to change the counter names in the Record modules to something
more obvious.
Let the run complete and then look at the output! Note that the demo runs for 1 million samples,
so the estimate ought to be very good.
a. 0
b. about 0.135
c. about 0.368
d. about 0.632
e. about 0.865
Solution: The answer is (c). In fact, when I ran my simulation, 367460 out of the 1 million fake
customers satisfied the inequality.
Of course, you can calculate the answer exactly (without using simulation):
