Total # of
Exam Takers
68
Min
Max
Mean
Median
47
100
83.4
88
25%
quantile
74.25
75 %
quantile
96
std
14.6
Bonus
avg.
12.17
1. (15) Suppose that X1 , ..., Xn are independent and identically distributed with expected value and
n
X
variance 2 . Let X = i=1n i . (Class Performance 50.8%)
(a) (3) What is E[3X1 X2 ]?
E[3X1 X2 ] = 3E[X1 ] E[X2 ] = 3 = 2
(b) (3) What is Var(X1 + 4 X2 + 2X1 )?
Var(X1 + 4 X2 + 2X1 ) = Var(3X1 X2 ) = 32Var(X1 ) +Var(X2 ) = 10 2
by independence.
ni=1 Xi
n E[Xi ] n
] = i=1
=
=
n
n
n
3. (25) Suppose you are running a restaurant having only one menu, fruit salad, in the Student
Center. You should order fruit salad every day 10pm after closing. Then, your supplier delivers
the ordered amount of fruit salad 5am next morning. Store hours is from 10am to 5pm every
day. The demand for the fruit salad for a day (10am - 5pm) has the following distribution. (Class
Performance 90%)
d
P(D=d)
20
0.1
25
0.2
30
0.4
35
0.2
40
0.1
The selling price of a fruit salad is $6, the buying price of a fruit salad is $1. Of course, leftover
fruit salad of a day cannot be used for future and you have to pay 50 cents per fruit salad for
disposal.
(a) (4) What is the optimal order-up-to quantity of fruit salad for a day?(a numerical answer is
expected)
Cp = 6,Cv = 1,Cs = 0.5.
(2points)F(y )
Cp Cv
61
5
=
=
0.769
Cp Cs
6 + 0.5 6.5
(e) (5) What is the optimal order-up-to quantity of fruit salad if the daily demand is exponentially
distributed with mean 30 units? Note that the pdf of exponential is f (x) = e x for x 0
where mean is 1/ . Its CDF is given as F(x) = 1 e x for x 0.
= 1/30.
1
30
y
1e
= 0.77, y = 30ln(0.23)
4. (50) Next months production at a manufacturing company will use a certain solvent for part
of its production process. You need to prepare solvent in prior and the demand of solvent is
random. Assume that there is an ordering cost of $1,500 incurred whenever an order for solvent
is placed and the solvent costs $20 per liter. Due to short product life cycle, unused solvent
cannot be used in following months. There will be a $10 disposal charge for each liter of solvent
left over at the end of the month. If there is a shortage of solvent, the production process is
seriously disrupted at a cost of $110 per liter short. Assume the demand is uniformly distributed
1
(continuous distribution) between 400 to 800 liters. (Note: D Uniform(a,b) has the pdf ba
and
xa
1
CDF ba , E[X] = 2 (a + b).)(Class Performance 79.6%)
(a) (5) What is the optimal ordering quantity (assume there is no initial inventory. A numerical
answer is expected.)
C f = 1500,Cv = 20,Co = 10,Cu = 110
Cu Cv
110 20
=
= 0.75
Cu +Co 110 + 10
y 400
= 0.75, y = 700.
800 400
(b) (5) Assume that the company currently has 400 liters in stock. What is the total expected cost
when no additional quantity is ordered?
F(y ) =
E[Cost] = Cu E[(D 400)+ ] +Co E[(400 D)+ ] = 110(E[D] 400) = 200 110 = 22000
since D 400.
(c) (8) Assume that the company currently has 400 liters in stock and additional quantity is
ordered. How many liters should be ordered if the answer for (a) is assumed to be 700
liters? (A numerical answer is exptected.)What is the total expected cost when the additional
quantity is ordered?
(3 points)order 700 - 400 = 300 liters.
E[Cost] = 1500 +Cv (y 400) +Cu E[(D 700)+ ] +Co E[(700 D)+ ]
= 1500 + 20 300 + 110E[(D 700)+ ] + 10E[(700 D)+ ]
= 1500 + 20 300 + 110
Z 800
(x 700)/400dx + 10
700
Z 700
400
(700 x)/400dx
(d) (7) Assume the total expected cost for no order is 22000 (question (b)) and the total
expected cost for order is 10000 (question (c)), should you order? What is the range of s
for the (s, S) policy?
(5 points) Since the cost for order is low, you should order. (2 points) And 400 < s < 700
since when x = 400, you should order.
(e) (15) Assume s > 400, give an equation that the parameter s of the (s, S) policy has to satisfy.
It can only involve s; it cannot have integrals, expectations or the variable S. For example,
2(s 400)2 = 8(400 + s)2 + 5(790 s) is acceptable. You may use the following quantities
to save time.
for any constant c,
for any constant c,
1 R 800
1
2
400 Rc (x c)dx = 800 (c 800)
c
1
1
2
400 400 (c x)dx = 800 (c 400)
1 R 800
400 700 (x 700)dx
1 R 700
400 400 (700 x)dx
= 12.5
= 112.5
(5points)E[ Cost for Order] = C f +Cv (y s) +Cu E[(D y )+ ] +Co E[(y D)+ ]
= 1500 + 20(700 s) + 110E[(D 700)+ ] + 10E[(700 D)+ ]
800
1
1
(x 700)dx + 10
400 700
400
= 1500 + 20(700 s) + 110 12.5 + 10 112.5
= 4000 + 20(700 s)
(5points)E[ Cost for no Order] = Cu E[(D s)+ ] +Co E[(s D)+ ]
= 110E[(D s)+ ] + 10E[(s D)+ ]
800
1
1
(x s)dx + 10
400 s
400
110
10
=
(s 800)2 +
(s 400)2
800
800
Z s
= 110
(s x)dx
400
Z 700
400
(700 x)dx
(g) (5) Identify the range of initial inventory level x such that the optimal order size is less or
equal to 500, assume (s = 900, S = 1200).
S x 500, x 700.
When x s = 900, the optimal order size is 0.