APPENDIX 3

Answers to Selected Problems

Chapter 2 SECTION 2.4

1 Linearly dependent.
SECTION 2.1
2 Linearly independent.

 
1 2 3
1 a  A  4 5 6 SECTION 2.5
7 8 9

 
12 12 12
1
 1 2 3

 
3 6 9 2 A
12 12 12
b 3A  12 15 18
21 24 27 3 A1 does not exist.
1
8 a 1B .
c A  2B is undefined. 100

 
1 4 7
SECTION 2.6
d AT  2 5 8
3 6 9 2 30.
0 1
2 
1
e BT  REVIEW PROBLEMS
1 2
1 Infinite number of solutions of the form x1  k  1,

 
4 6 x2  3  k, x3  k.
f AB  10 15
 T   0.20  T 
Ut1 0.75 0 Ut
3
16 24 t1 0.90 t

g BA is undefined. 4 x1  0, x2  1.

   
y1 0.50 0 0.10 x1 13 Linearly independent.
2 y2  0.30 0.70 0.30 x2 14 Linearly dependent.
y3 0.20 0.30 0.60 x3
15 a Only if a, b, c, and d are all nonzero will rank A 
4. Thus, A1 exists if and only if all of a, b, c, and d
SECTION 2.2 are nonzero.
b Applying the Gauss-Jordan method, we find if a, b,

       
1 1 4 1 1 4 c, and d are all nonzero,
x
1 2 1 1  6 or 2 1 6

 
x 1 0 0 0
1 3 2 8 1 3 8 a
1
0  0 0
A1  b
0 0 1 0
SECTION 2.3 c
0 0 0 1d
1 No solution. 18 4.
2 Infinite number of solutions of the form x1  2  2k,
x2  2  k, x3  k.
3 x1  2, x2  1.
Chapter 3 s.t. 8(x1  x4  x5  x6  x7)  4(x8  x11  x12
 x13  x14)  88 (Sunday)
SECTION 3.1
s.t. 8(x1  x2  x5  x6  x7)  4(x8  x9  x12
1 max z  30x1  100x2  x13  x14)  136 (Monday)
1 s.t. x1  x2  7 (Land constraint) s.t. 8(x1  x2  x3  x6  x7)  4(x8  x9  x10
4x1  10x2  40 (Labor constraint)  x13  x14)  104 (Tuesday)
1 s.t. 10x1 + 10 x2  30 (Government constraint) s.t. 8(x1  x2  x3  x4  x7)  4(x8  x9  x10
x1, x2  0  x11  x14)  120 (Wednesday)
2 No, the government constraint is not satisfied. s.t. 8(x1  x2  x3  x4  x5)  4(x8  x9  x10
b No, the labor constraint is not satisfied.
 x11  x12)  152 (Thursday)
c No, x2  0 is not satisfied.
s.t. 8(x2  x3  x4  x5  x6)  4(x9  x10  x11
SECTION 3.2  x12  x13)  112 (Friday)

1 z  $370, x1  3, x2  2.8. s.t. 8(x3  x4  x5  x6  x7)  4(x10  x11  x12

3 z  $14, x1  3, x2  2.  x13  x14)  128 (Saturday)

4 a We want to make x1 larger and x2 smaller, so we 20(x8  x9  x10  x11  x12  x13  x14)
move down and to the right.  0.25(136  104  120  152  112  128
b We want to make x1 smaller and x2 larger, so we
88)
move up and to the left.
b We want to make both x1 and x2 smaller, so we move (The last constraint ensures that part-time labor
down and to the left. will fulfill at most 25% of all labor requirements)
All variables  0
SECTION 3.3 3 Let x1  number of employees who start work on Sunday
1 No feasible solution. and work five days, x2  number of employees who start
work on Monday and work five days, . . . , x7  number of
2 Alternative optimal solutions. employees who start work on Saturday and work five days.
3 Unbounded LP. Also let o1  number of employees who start work on
Sunday and work six days, . . . , o7  number of employees
who start work on Saturday and work six days. Then the
SECTION 3.4 appropriate LP is
1 For i  1, 2, 3, let xi  tons of processed factory i min z  250(x1  x2      x7)
waste. Then the appropriate LP is min z  + 312(o1  o2      o7)
min z  15x1  10x2  20x3 s.t. x1  x4  x5  x6  x7  o1  o3  o4
s.t. 0.10x1  0.20x2  0.40x3  30 (Pollutant 1)  o5  o6  o7  11 (Sunday)
s.t. 0.45x1  0.25x2  0.30x3  40 (Pollutant 2) s.t. x1  x2  x5  x6  x7  o1  o2  o4
x1, x2, x3  0  o5  o6  o7  17 (Monday)
It is doubtful that the processing cost is proportional to the s.t. x1  x2  x3  x6  x7  o1  o2  o3
amount of waste processed. For example, processing 10
tons of waste is probably not 10 times as costly as processing  o5  o6  o7  13 (Tuesday)
1 ton of waste. The Divisibility and Certainty Assumptions s.t. x1  x2  x3  x4  x7  o1  o2  o3
seem reasonable.
 o4  o6  o7  15 (Wednesday)
SECTION 3.5 s.t. x1  x2  x3  x4  x5  o1  o2  o3

1 Let x1  number of full-time employees (FTE) who  o4  o5  o7  19 (Thursday)

start work on Sunday, x2  number of FTE who start work s.t. x2  x3  x4  x5  x6  o1  o2  o3
on Monday, . . . , x7  number of FTE who start work on
Saturday; x8  number of part-time employees (PTE) who  o4  o5  o6  14 (Friday)
start work on Sunday, . . . , x14  number of PTE who start s.t. x3  x4  x5  x6  x7  o2  o3  o4
work on Saturday. Then the appropriate LP is
 o5  o6  o7  16 (Saturday)
min z  15(8)(5)(x1  x2      x7)
All variables  0
min z   10(4)(5)(x8  x9      x14)

APPENDIX 3 Answers to Selected Problems 1371

SECTION 3.6 1 max z   x3  2o1  3o2
2 NPV of investment 1 1 max z  s.t. o1  2x1  x2

5 7 9 1 max z  s.t. o2  3x1  3x2  2x3

= 6    2  3  $2.00.
1.1 (1.1) (1.1) 1 max z  s.t. o1  200
NPV of investment 2 1 max z  s.t. o2  300
3 9 7 g2  3x3  x1  3x2 (Gas 2 production)
= 8    2  3  $1.97.
1.1 (1.1) (1.1) x1  x2  x3  100 (100 hours per week
Let x1  fraction of investment 1 that is undertaken and of cracker time)
x2  fraction of investment 2 that is undertaken. If we All variables  0
measure NPV in thousands of dollars, we want to solve the
following LP: 5 Let A  total number of units of A produced

max z  2x1  1.97x2 B  total number of units of B produced

s.t. 6x1  8x2  10 CS  total number of units of C produced (and

sold)
5x1  3x2  7
AS  units of A sold
x1  3x2  1
BS  units of B sold
x2  1
max z  10AS  56BS  100CS
All variables  0
s.t. A  2B  3C  40
The optimal solution to this LP is x1  1, x2  0.5, z 
$2,985. A  AS  2B
B  BS  CS
SECTION 3.7 All variables  0
1 z  $2,500, x1  50, x2  100.
SECTION 3.10
SECTION 3.8 1 Let xt  production during month t and it  inventory
at end of month t.
1 Let ingredient 1  sugar, ingredient 2  nuts, ingredient
3  chocolate, candy 1  Slugger, and candy 2  Easy 1 min z  5x1  8x2  4x3  7x4
Out. Let xij  ounces of ingredient i used to make candy j.
1 min z =  2i1  2i2  2i3  2i4  6i4
(All variables are in ounces.) The appropriate LP is
1 s.t. i1  x1  50
max z  25(x12  x22  x32)  20(x11  x21  x31)
1 s.t. i2  i1  x2  65
s.t. x11  x12  100 (Sugar constraint)
1 s.t. i3  i2  x3  100
x21  x22  20 (Nuts constraint)
1 s.t. i4  i3  x4  70
x31  x32  30 (Chocolate constraint)
All variables  0
x22  0.2(x12  x22  x32)
x21  0.1(x11  x21  x31) SECTION 3.11
x31  0.1(x11  x21  x31)
3 Let A  dollars invested in A, B  dollars invested in
All variables  0 B, c0  leftover cash at time 0, c1  leftover cash at time
1, and c2  leftover cash at time 2. Then a correct
SECTION 3.9 formulation is
max z  c2  1.9B
1 Let x1  hours of process 1 run per week
s.t. A  c0  10,000
1 Let x2  hours of process 2 run per week
(Time 0 available  time 0 invested)
1 Let x3  hours of process 3 run per week
0.2A  c0  B  c1
1 Let g2  barrels of gas 2 sold per week
(Time 1 available  time 1 invested)
1 Let o1  barrels of oil 1 purchased per week
1.5A  c1  c2
1 Let o2  barrels of oil 2 purchased per week
(Time 2 available  time 2 invested)
1 max z  9(2x1)  10g2  24(2x3)  5x1  4x2
All variables  0

1372 APPENDIX 3 Answers to Selected Problems

The optimal solution to this LP is B  c0  $10,000, A  min z  100x1  90x2  120x3  80x4
c1  c2  0, and z  $19,000. Notice that it is optimal to
s.t. x1  x2  100
wait for the “good” investment (B) even though leftover
cash earns no interest. (Farm 1 land)
x3  x4  100
SECTION 3.12
(Farm 2 land)
2 Let JAN1  number of computers rented at beginning 500x1  650x3  7,000
of January for one month, and so on. Also define IJAN 
number of computers available to meet January demand, (Corn requirement)
and so on. The appropriate LP is 400x2  350x4  11,000
min z  200(JAN1  FEB1  MAR1  APR1 (Wheat requirement)
min z  + MAY1  JUN1)  350(JAN2  FEB2 x1, x2, x3, x4  0
min z  + MAR2  APR2  MAY2  JUN2) 9 Let x1  units of process 1, x2  units of process 2, and
min z  + 450(JAN3  FEB3  MAR3  APR3) x3  modeling hours hired. Then a correct formulation is
min z  + MAY3  JUN3)  150MAY3 max z  5(3x1  5 x2)  3(x1  2x2)
min z   300 JUN3  175JUN2  2(2x1  3x2)  100x3
s.t. IJAN  JAN1  JAN2  JAN3 s.t. x1  2x2  20,000 (Limited labor)
IFEB  IJAN  JAN1  FEB1  FEB2  FEB3 2x1  3x2  35,000 (Limited chemicals)
s.t. IMAR  IFEB  JAN2  FEB1  MAR1 3x1  5x2  1,000  200x3
 MAR2  MAR3 (Perfume production  perfume demands)
IAPR  IMAR  FEB2  MAR1  JAN3 x1, x2, x3  0
 APR1  APR2  APR3 17 Let OT  number of tables made of oak, OC  number
of chairs made of oak, PT  number of tables made of pine,
IMAY  IAPR  FEB3  MAR2  APR1 and PC  number of chairs made of pine. Then the correct
s.t. IMAY   MAY1  MAY2  MAY3 formulation is
IJUN  IMAY  MAR3  APR2  MAY1 max z  40(OT  PT)  15(OC  PC)
 JUN1  JUN2  JUN3 s.t. 17(OT)  5(OC)  150
IJAN  9 (Use at most 150 board ft of oak)
IFEB  5 30PT  13PC  210
IMAR  7 (Use at most 210 board ft of pine)
IAPR  9 OT, OC, PT, PC  0
IMAY  10 18 Let school 1  Cooley High, and school 2  Walt
Whitman High. Let Mij  number of minority students who
IJUN  5 live in district i who will attend school j, and let NMij 
All variables  0 number of nonminority students who live in district i who
will attend school j. Then the correct LP is
REVIEW PROBLEMS min z  (M11  NM11)  2(M12  NM12)

2 Let x1  number of chocolate cakes baked and x2   2(M21  NM21)  (M22  NM22)
number of vanilla cakes baked. Then we must solve  (M31  NM31)  (M32  NM32)
max z  x1  12 x2 s.t. M11  M12  50
1 2
s.t.  x1
3
  x2
3
8 M21  M22  50
4x1  x2  30 M31  M32  100
x1, x2  0 NM11  NM12  200
$69 36 66
The optimal solution is z  ,
7
x1  ,
7
x2  .
7 NM21  NM22  250
8 Let x1  acres of farm 1 devoted to corn, x2  acres of NM31  NM32  150
farm 1 devoted to wheat, x3  acres of farm 2 devoted to
corn, x4 acres of farm 2 devoted to wheat. Then a correct
formulation is

APPENDIX 3 Answers to Selected Problems 1373

For school 1, we obtain the following blending constraints: s.t. X16  X26  X36  X46  X56  X67  9
M11  M21  M31 (Friday constraint)
0.20  
M11  M21  M31  NM11  NM21  NM31 All variables  0
 0.30 49 Let Xij  money invested at beginning of month i for
This yields the following two LP constraints: a period of j months. After noting that for each month
(money invested)  (bills paid)  (money available), we
0.8M11  0.8M21  0.8M31  0.2NM11 obtain the following formulation:
 0.2NM21  0.2NM31>  0 max z  1.08X14  1.03X23  1.01X32  1.001X41
0.7M11  0.7M21  0.7M32  0.3NM11 s.t. X11  X12  X13  X14  600  400  400
 0.3NM21  0.3NM31  0 (Month 1)
For school 2, we obtain the following blending constraints: s.t. X21  X22  X23  500  1.001X11  800
M12  M22  M32 (Month 2)
0.20  
M12  M22  M32  NM12  NM22  NM32 s.t. X31  X32  500  1.01X12  1.001X21  300
 0.30 (Month 3)
This yields the following two LP constraints: s.t. X41  250  1.03X13  1.01X22  1.001X31  300
0.8M12  0.8M22  0.8M32  0.20NM12 (Month 4)
 0.20NM22  0.20NM32  0 All variables  0
0.7M12  0.7M22  0.7M32  0.30NM12 53 Let T1  number of type 1 turkeys purchased
 0.30NM22  0.30NM32  0 53 Let T2  number of type 2 turkeys purchased
We must also ensure that each school has between 300 and D1  pounds of dark meat used in cutlet 1
500 students. Thus, we also need the following constraints:
W1  pounds of white meat used in cutlet 1
300  M11  NM11  M21  NM21  M31  NM31
D2  pounds of dark meat used in cutlet 2
 500
W2  pounds of white meat used in cutlet 2
300  M12  NM12  M22  NM22  M32  NM32
Then the appropriate formulation is
 500
max z  4(W1  D1)  3(W2  D2)  10T1  8T2
To complete the formulation, add the sign restrictions that
all variables are  0. s.t. W1  D1  50 (Cutlet 1 demand)
47 For i j, let Xij  number of workers who get off days s.t. W2  D2  30 (Cutlet 2 demand)
i and j of week (day 1  Sunday, day 2  Monday, . . . , day s.t. W1  W2  5T1  3T2 (Don’t use more white
7  Saturday). meat than you have)
max z  X12  X17  X23  X34  X45  X56  X67 D1  D2  2T1  3T2 (Don’t use more dark meat
s.t. X17  X27  X37  X47  X57  X67  2 than you have)
(Saturday constraint) W1/(W1  D1)  0.7 or 0.3W1  0.7D1
s.t. X12  X13  X14  X15  X16  X17  12 W2/(W2  D2)  0.6 or 0.4W2  0.6D2
(Sunday constraint) T1, T2, D1, W1, D2, W2  0
s.t. X12  X23  X24  X25  X26  X27  12
(Monday constraint) Chapter 4
s.t. X13  X23  X34  X35  X36  X37  6
SECTION 4.1
(Tuesday constraint)
1 max z  3x1  2x2
s.t. X14  X24  X34  X45  X46  X47  5
1 s.t. 2x1  x2  s1 + s2  s3  100
(Wednesday constraint)
x1  x2 + s1 + s2  80
s.t. X15  X25  X35  X45  X56  X57  14
x1  s3  40
(Thursday constraint)

1374 APPENDIX 3 Answers to Selected Problems

3 min z  3x1  x2 Set 1  {x1, x2, s1}
1 s.t. x1 + x2  e1 + s2  3 Set 2  {x1, x2, s2}
1 s.t. x1  x2  s2  4 Set 3  {x1, x2, s3}
1 s.t. 2x1  x2 3
SECTIONS 4.12 AND 4.13
SECTION 4.4 1 z  1, x1  x2  0, x3  1.
1 From Figure 2 of Chapter 3, we find the extreme points 4 Infeasible LP.
of the feasible region.
Point Basic Variables SECTION 4.14

H  (0, 0) s1  100, s2  80, s3  40 1 Let it  i

t  it be the inventory position at the end of
E  (40, 0) x1  40, s1 month t. For each constraint in the original problem, replace
 20, s2  40
it by i
t  it. Also add the sign restrictions i
t  0 and it  0.
F  (40, 20) x1  40, x2  20, s2  20 To ensure that demand is met by the end of quarter 4, add
G  (20, 60) x1  20, x2  60, s3  20 constraint i 4  0. Replace the terms involving it in the
D  (0, 80) x2  80, s1  20, s3  40 objective function by
(100i
1  100i1  100i
2  100i2
SECTION 4.5
 100i
3  110i3  100i
4  110i4)
1 z  180, x1  20, x2  60. 2 z  5, x1  1, x2  3.
32 10
2 z ,
3
x1  ,
3
x2  43.
SECTION 4.16
SECTION 4.6
2 Let xi  number of lots purchased from supplier i. The
1 z  5, x1  0, x2  5. appropriate LP is
min z  10s   
1  6s 2  4s 3  s 4
4.7 60x1  50x2  40x3  s1  s
1  5,000
SECTION s.t.
2 Solution 1: z  6, x1  0, x2  1; solution 2: z  6, (Excellent chips)
x1  5167, x2  4157. By averaging these two solutions, we 20x1  35x2  20x3  s2  s
s.t. 2  3,000
obtain solution 3: z  6, x1  2187 , x2  3117 .
(Good chips)
SECTION 4.8 s.t. 20x1  15x2  40x3  s3  s
3  1,000

(Mediocre chips)
1 x1  4,999, x2  5,000 has z  10,000.
s.t. 400x1  300x2  250x3  s4  s4  28,000
SECTION 4.10 (Budget constraint)
1 a Both very small numbers (for example, 0.000003) All variables  0
and large numbers (for example, 3,000,000) appear in
the problem. REVIEW PROBLEMS
b Let x1  units of product i produced (in millions).
If we measure our profit in millions of dollars, the LP 4 Unbounded LP.
becomes 5 z  6, x1  0, x2  3.
1 max  6x1  4x2  3x3 6 Infeasible LP.
1 s.t. 4x1  3x2  2x3  3 (Million labor hours) 8 z  12, x1  x2  2.
1 s.t. 3x1  2x2  x3  2 (lb of pollution) 10 Four types of furniture.
1 s.t. 3x1, x2, x3  0 15 z 17
, x2  32, x4  12.
2
17 a c  0 and b  0.
SECTION 4.11
b b  0 and c  0. Also need a2 > 0 and/or a3 > 0
1 z  16, x1  x2  2. The point where all three constraints to ensure that when x1 is pivoted in, a feasible solution
are binding (x1  x2  2) corresponds to the following results. If only a3 > 0, then we need b to be strictly
three sets of basic variables: positive.

APPENDIX 3 Answers to Selected Problems 1375

 c c 0, a2  0, a3  0 ensures that x1 can be s.t. C3  200R3
made arbitrarily large and z will become arbitrarily
large. s.t. C4  200R4, I4  0

20 Let ct  net number of drivers hired at the beginning B1, B2, B3, B4  2
of the year t. Then ct  ht  ft, where ht  number of drivers All variables  0
hired at beginning of year t, and ft  number of drivers fired
at beginning of year t. Also let dt  number of drivers after
drivers have been hired or fired at beginning of year t. Then Chapter 5
a correct formulation is (cost in thousands of dollars)
min z  10(d1  d2  d3  d4  d5) SECTION 5.1
 2(f1  f2  f3  f4  f5) 1 Decision variables remain the same. New z-value is
 4(h1  h2  h3  h4  h5) $210.
s.t. d1  50  h1  f1 4 a 530  c1  350.
c 4,000,000  HIW  84,000,000; x1  3.6  0.15,
s.t. d2  d1  h2  f2
x2  1.4  0.025.
s.t. d3  d2  h3  f3 f $310,000.
s.t. d4  d3  h4  f4
s.t. d5  d4  h5  f5 SECTION 5.2

s.t. d1  60, d2  70, d3  50, d4  65, 1 a $3,875.

s.t. d5  75 b Decision variables remain the same. New z-value is
$3,750.
All variables  0 c Solution remains the same.
26 Let 3 a Still 90¢.
Rt  robots available during quarter t
b 95¢.
(after robots are bought or sold for the quarter)
c 95¢.
Bt  robots bought during quarter t d Still 90¢.
St  robots sold during quarter t e 82.5¢.
I
t  cars in inventory at end of quarter t f 30¢ or less.
g 22.5¢ or less.
Ct  cars produced during quarter t
It  backlogged demand for cars at end of quarter t SECTION 5.3
Then a correct formulation is
3 2.5¢.
min z  500(R1  R2  R3  R4)
4 $2.
 200(I
1  I
2  I
3  I
4)
5 Buy raw material, because it will reduce cost by $6.67.
 5,000(B1  B2  B3  B4)
 3,000(S1  S2  S3  S4) SECTION 5.4
 300(I 1  I2  I 3  I4) 3 See Figures 1–4.
s.t. R1  2  B1  S1
s.t. R2  R1  B2  S2 FIGURE 1

s.t. R3  R2  B3  S3
s.t. R4  R3  B4  S4
I
1  I 1  C1  600
I
2  I2  I
1  I 1  C2  800
I
3  I 3  I
2  I2  C3  500
I
4  I4  I
3  I 3  C4  400
s.t. R4  2
s.t. C1  200R1
s.t. C2  200R2

1376 APPENDIX 3 Answers to Selected Problems

FIGURE 2 REVIEW PROBLEMS

1 a $1,046,667.
b Yes.
c $33.33.
b $333.33.
7 a Decision variables remain unchanged. New z-value
is $1,815,000.
b Pay $0 for an additional 100 board ft of lumber. Pay
$1,350 for an additional 100 hours of labor.
c $1,310,000.
d $1,665,000.

Chapter 6

SECTION 6.1

1 Decision variables remain the same. New z-value is

$210.
FIGURE 3
SECTION 6.2

1 z  4s1  5s2  28
x1  s1  s2  6
x2  s1  2s2  10

SECTION 6.3

3 x1  2, x2  0, x3  8, z  280 (same as original

solution).
5 Home computer tables should not be produced.
6 a Profit for candy bar 1  6¢. If type 1 candy bar
earns 7¢ profit, new optimal solution is z  $3.50,
x1  50, x2  x3  0.
b 5¢  candy bar 2 profit  15¢. If candy bar 2 profit
is 13¢, decision variables remain the same, but profit is
now $4.50.
FIGURE 4 100
c    sugar  100.
3
d z  $3.40, x1  0, x2  20, x3  40. If 30 oz of
sugar is available, current basis is no longer optimal,
and problem must be solved again.
e Make type 1 candy bars.
f Make type 4 candy bars.
8 a $16,667.67  comedy cost  $350,000.
b 4 million  HIW  84 million. For 40 million HIW
exposures, new optimal solution is x1  5.4,
x2  1.1, z  $350,000.
c Advertise on news program.

SECTION 6.4

1 Yes.
2 No.
4 Yes.

APPENDIX 3 Answers to Selected Problems 1377

SECTION 6.5 SECTION 6.8
1 min w  y1  3y2  4y3 2 b New z-value  $3.40.
s.t. y1  y2  y3  2 c New z-value  $2.60.
d Since current basis is no longer optimal, the current
y1  y2  2y3  1
shadow prices cannot be used to determine the new
y1, y2, y3  0 z-value.
2 max z  4x1  x2  3x3 5 b Skilled labor shadow price  0, unskilled labor
shadow price  0, raw material shadow price  15, and
s.t. 2x1  x2  x3  1
product 2 constraint shadow price  5.
x1  x2  2x3  1 We would be willing to pay $0 for an additional hour
of either type of labor. We would pay up to $15 for an extra
x1, x2, x3  0
unit of raw material. Reducing the product 2 marketing
3 min w  5y1  7y2  6y3  4y4 requirement by 1 unit will save the company $5.
s.t. y1  2y2  y4  4 c b3  5, so new z-value  435  5(15)  $510.
d Since shadow price of each labor constraint is zero,
1 s.t. y1  y2  2y3  1 the optimal z-value remains unchanged.
1 s.t. y1  y2  2y3  y4  2 e For a 5-unit requirement, b4  2. Thus, new
1 s.t. y1  y1, y2  0; y3  0; y4 urs z-value  435  2(5)  $425. For a 2-unit requirement,
b4  1. Thus, new z-value  435  (1)(5) 
4 max z  6x1  8x2 $440.
s.t. x1  x2  4 6 a If purchased at the given price of $1, an extra unit
1 s.t. 2x1  x2  2 of raw material increases profits by $2.50. Thus, the firm
would be willing to pay up to 1  2.5  $3.50 for an
1 s.t. 2x1  2x2  1 extra unit of raw material.
1 s.t. x1  0; x2 urs b Both labor constraints are nonbinding. All we can
say is that if an additional hour of skilled labor were
available at $3/hour, we would not buy it, and if an
SECTION 6.7 additional hour of unskilled labor were available at
1 a min w  100y1  80y2  40y3 $2/hour, we would not buy it.

s.t. 2y1  y2  y3  3 7 a New z-value  $380,000.

b New z-value  $290,000.
y1  y2 2
1 a s.t. y1 , y2, y3  0 SECTION 6.9
b and c y1  1, y2  1, y3  0, w  180. Observe
that this solution has a w-value that equals the optimal 1 The current basis is no longer optimal. We should make
primal z-value. Since this solution is dual feasible, it computer tables, because they sell for $35 each and use only
must be optimal (by Lemma 2) for the dual. $30 worth of resources.
2 a min w  3y1  2y2  y3 2 a Current basis remains optimal if type 1 profit  6¢.
s.t. y1  y3  2
SECTION 6.10
y1  y2  1
1 a min w  600y1  400y2  500y3
y1  y2  y3  1
b y1  coefficient of s1 in optimal row 0  0 s.t. 4y1  y2  3y3  6
b y2  (coefficient of e2 in optimal row 0) 1 a s.t. 9y1  y2  4y3  10
b y2  1 1 a s.t. 7y1  3y2  2y3  9
b y3  coefficient of a3 in optimal row 0  M 1 a s.t. 10y1  40y2  y3  20
b y2  2 1 a s.t. 10y1 y1, y2, y3, y4  0
2,800 22 2
b w  , y1  , y2  , y3  0.
b y2  coeffiOptimal w-value  0. 3 5 15

9 Dual is max w  28y1  24y2 SECTION 6.11

s.t. 7y1  2y2  50
1 z  9, x1  0, x2  14, x3  9.
2y1  12y2  100
2 a The current solution is still optimal.
y1, y2  0 b The LP is now infeasible.
Optimal dual solution is w  $320,000, y1  5, y2  7.5.

1378 APPENDIX 3 Answers to Selected Problems

 c The new optimal solution is z  10, x1  1, 4 a New z-value  32,540  10(88)  $33,420.
x2  4. Decision variables remain the same.
b Can’t tell, since allowable increase is 1.
SECTION 6.12 c $0.
d 32,540  (2)(20)  $32,580.
4 Only HPER is inefficient.
e Produce jeeps.

REVIEW PROBLEMS 8 a New z-value  $266.20.

b New z-value  $270.70. Decision variables remain
1 a min w  6y1  3y2  10y3 the same.
1 a s.t. y1  y2  2y3  4 c $12.60.
d 20¢.
1 a s.t. 2y1  y2  y3  1
e Produce product 3.
1 a s.y1 urs; y2  0; y3  0
17 z  16, x1  8, x2  0.
Optimal dual solution is w  538, y1  23, y2  0, y3  73.
20 Optimal primal solution: z  13, x1  1, x2  x3  0,
b 9  b3  12. If b3  11, the new optimal solution
x4  2. Optimal dual solution: w  13, y1  1, y2  1.
is z  635, x1  136, x2  13.
21 a c1  3.
2 c1  12.
b c2  43.
3 a min w  6y1  8y2  2y3 c 0  b1  9.
3 a s.t. y1  6y2 + y3  5 d b2  10.
3 a s.t. y1  y3  1 28 LP 2 optimal solution: z  550, x1  0.5, x2  5.
100
3 a s.t. y1  y2  y3  2 Optimal solution to dual of LP 2: w  550, y1  y2  3
.

3 a s.t. y1, y2, y3  0 36 b2  3.

5 7
Optimal dual solution is w  9, y1  0, y2  ,
6
y3  .
6
b 0  c1  6.
c c2  76.

Chapter 7
SECTION 7.1
1

APPENDIX 3 Answers to Selected Problems 1379

3

1380 APPENDIX 3 Answers to Selected Problems

7 This is a maximization problem, so number in each cell
is a revenue, not a cost.

12 a Replace the M’s by incorporating a backlogging Laroach in month 1.

cost. For example, month 3 regular production can
be used to meet month 1 demand at a cost of 400 
2(30)  $460.
b Add a supply point called “lost sales,” with cost of
shipping a unit to any month’s demand being $450.
Supply of “lost sales” supply point should equal total
demand. Then adjust dummy demand point’s demand
to rebalance the problem.
c A shipment from month 1 production to month 4
demand should have a cost of M.
d For each month, add a month i subcontracting
supply point, with a supply of 10 and a cost that is $40
more than the cost for the corresponding month i
regular supply point. Then adjust the demand at the
dummy demand point so that the problem is balanced.
SECTIONS 7.2 AND 7.3
The optimal solution to Problem 1 of Section 7.1 is to ship 10
units from warehouse 1 to customer 2, 30 units from warehouse
1 to customer 3, and 30 units from warehouse 2 to customer 1.
The optimal solution to Problem 5 of Section 7.1 is to
buy 4 gallons from Daisy in month 2 and 3 gallons from
In Problem 7 of Section 7.1, Cliff gets 20,000 acres at
site 1 and 20,000 acres at site 2. Blake gets 80,000 acres at
site 2. Alexis gets 80,000 acres at site 1.
SECTION 7.4
2 Current basis remains optimal if c34  7.
4 New optimal solution is x12  12, x13  23, x21  45,
x23  5, x32  8, x34  30, and z  1,020  2(3)  2(10)
 994.
SECTION 7.5
1 Person 1 does job 2, person 2 does job 1, person 3 does
no job, person 4 does job 4, and person 5 does job 3.
8 a Company 1 does route 1, company 2 does route 2,
company 3 does route 3, and company 4 does route 4.
b Company 3 does routes 3 and 1, company 2 does
routes 2 and 4.

SECTION 7.6

1 a

APPENDIX 3 Answers to Selected Problems 1381

 b

REVIEW PROBLEMS

3 Meet January demand with 30 units of January from Bi to Bj means one black student from district i goes
production. Meet February demand with 5 units of January to school in district j. The costs of M ensure that shipments
production, 10 units of February production, and 15 units of from Wi to Bj or Bi to Wj cannot occur (table on next page).
March production. Meet March demand with 20 units of
8 Optimal solution is z  1,580, x11  40, x12  10, x13
March production.
 10, x22  50, x32  10, x34  30.
4 Maid 1 does the bathroom, maid 2 straightens up, maid
13 Optimal solution is z  98, x13  5, x21  3, x24  7,
3 does the kitchen, maid 4 gets the day off, and maid 5
x32  3, x33  7, x34  5.
vacuums.
25 Sell painting 1 to customer 1, painting 2 to customer
7 Shipping 1 unit from Wi to Wj means one white student
2, painting 3 to customer 3, and painting 4 to customer 4.
from district i goes to school in district j. Shipping 1 unit

1382 APPENDIX 3 Answers to Selected Problems

Chapter 8

SECTION 8.2

2 1–2–5 (length 14).

3

M  large number to prevent shipping a unit through a 1 s.t. x0  xso,1  xso,2 (Node so)
nonexistent arc.
1 s.t. xso,1  x13  x12 (Node 1)
5 Replace the car at times 2, 4, and 6. Total cost  $14,400.
1 s.t. x12  x32  xso,2  x24 (Node 2)

SECTION 8.3 1 s.t. x13  x32  x3,si (Node 3)

1 s.t. x24  x4,si (Node 4)
1 max z  x0
1 s.t. x3,si  x4,si  x0 (Node si )
1 s.t. xso,1  6, xso,2  2, x12  1, x32  3,
1 s.t. All variables  0
1 s.t. x13  3, x3,si  2, x24  7, x4,si  7

APPENDIX 3 Answers to Selected Problems 1383

Maximum flow  6. Cut associated with V
 {2, 3, 4, si} SECTION 8.4
has capacity 6.
4 a See Figure 7.
2 max z  x0
1 s.t. xso,1  2, x12  4, x1,si  3, x2,si  2,
FIGURE 7
1 s.t. x23  1, x3,si  2, xso,3  1
1 s.t. x0  xso,1  xso,3 (Node so)
1 s.t. xso,1  x1,si  x12 (Node 1)
1 s.t. x12  x23  x2,si (Node 2)
1 s.t. x23  xso,3  x3,si (Node 3)
1 s.t. x1,si  x2,si  x3,si  x0 (Node si )
1 s.t. All variables  0
Maximum flow  3. Cut associated with V
 {1, 2, 3, si}
has capacity 3.
6 See Figure 5. An arc of capacity 1 goes from each
package type node to each truck node. If maximum flow  b Critical path is A–C–G (project duration is 14 days).
21, all packages can be delivered. c Start project by June 13.
d min z  x8  x1

FIGURE 5 d s.t. x2  x1  3
d s.t. x3  x2  6
d s.t. x4  x2  2
d s.t. x5  x4  3
d s.t. x5  x3  1
d s.t. x6  x5  1.5
d s.t. x8  x6  2
d s.t. x8  x7  3(x7  x3  2)
d s.t. x8  x2  3
d s.t. x8  x3  5
All variables urs
5 a See Figure 8. A–B–E–F–G and A–B–C–G are critical
paths. Duration of project is 26 days.

FIGURE 8
7 See Figure 6. If maximum flow  4, then all jobs can
be completed.

FIGURE 6

Activity Total Float Free Float

A 0 0
B 0 0
C 0 0
D 8 8
E 0 0
F 0 0
G 0 0

1384 APPENDIX 3 Answers to Selected Problems

 b min z  30A  15B  20C  40D Arc Shipping Cost
b min z  20E  30F  40G Bos.–Chic. 800  80  $1,880
b s.t. x2  x1  5  A Bos.–Aus. 800  220  $1,020
b s.t. x3  x2  8  B Bos.–L.A. 800  280  $1,080
Ral.–Chic. 900  100  $1,000
b s.t. x4  x3  4  E
Ral.–Aus. 900  140  $1,040
b s.t. x5  x3  10  C Ral.–L.A. 900  170  $1,070
b s.t. x5  x4  6  F Chic.–Aus. $40
b s.t. x6  x3  5  D Chic.–L.A. $50
b s.t. x6  x5  3  G Problem is balanced, so no dummy point is needed.
b s.t. x6  x1  20 City Net Outflow
b s.t. A  2, B  3, C  1, D  2, E  2, Boston 400
b s.t. F  3, G  1 Raleigh 300
Chicago 0
b s.t. A, B, C, D, E, F, G  0
L.A. 400
All other variables urs Austin 300
8 b From the LP output, we find the critical path 5
1–2–3–4–5–6. This implies that activities A, B, E, F, and
G are critical. (Since 1–2–3–5–6 is also a critical path, Arc Unit Cost
activity C is also a critical activity, but the LP does not
give us this information.) S.D.–Dal. $420
S.D.–Hous. $100
SECTION 8.5 L.A.–Dal. $300
L.A.–Hous. $110
1 min z  4x12  3x24  2x46  3x13 S.D.–Dummy $0
1 min z   3x35  2x25  2x56 L.A.–Dummy $0
1 s.t. x12  x13  1 (Node 1) Dal.–Chic. 700  550  $1,250
Dal.–N.Y. 700  450  $1,150
1 s.t. x12  x24  x25 (Node 2)
Hous.–Chic. 900  530  $1,430
1 s.t. x13  x35 (Node 3) Hous.–N.Y. 900  470  $1,370
1 s.t. x24  x46 (Node 4) Net Outflow
1 s.t. x25  x56 (Node 5) City (100,000 barrels/day)

1 s.t. x46  x56  1 (Node 6) San Diego 5

L.A. 4
xij  0
Dallas 0
If xij  1, the shortest path from node 1 to node 6 contains Houston 0
arc (i, j); if xij  0, the shortest path from node 1 to node 6
does not contain arc (i, j). Chicago 4
N.Y. 3
4 a See Figure 9. All arcs have infinite capacity.
Dummy 2

FIGURE 9 SECTION 8.6

2 The MST consists of the arcs (1, 3), (3, 5), (3, 4), and
(3, 2). Total length of MST is 15.

SECTION 8.7

1 c z  8, x12  x25  x56  1, x13  x24  x35 

x46  0.
3 z  590, x12  20, x24  20, x34  2, x35  2, x13 
12, x23  0, x25  0, x45  0.

APPENDIX 3 Answers to Selected Problems 1385

REVIEW PROBLEMS s.t. 3x1  x2  3x3  3x4  3x5  x6  2x7  10
(Shooting)
1 a N.Y.–St. Louis–Phoenix–L.A. uses 2,450 gallons of
fuel. s.t. x1  3x2  2x3  3x4  3x5  2x6  2x7  10
(Rebounding)
2 a See Figure 10.
s.t. x6  x3  1
s.t. x4  x5  2  2y
FIGURE 10 (If x1 0, then x4  x5  2)
s.t. x1  2(1  y)
s.t. x2  x3  1
.x1, x2, . . . , x7, y all 0–1 variables
3 Let x1  units of product 1 produced
3 Let x2  units of product 2 produced

0
1 if any product i is ordered
3 Let yi 
otherwise
Then the appropriate IP is
max z  2x1  5x2  10y1  20y2
s.t. 3x1  6x2  120
2 max z  x0
s.t. x1 3  6x2 40y1
2 s.t. x12  x13  x0 (Node 1)
s.t. 3x1  6x2  20y2
2 s.t. x12  x24  x25 (Node 2)
x1, x2  0; y1, y2  0 or 1
2 s.t. x13  x34  x35 (Node 3)
0
1 if calculus is taken
2 s.t. x24  x34  x46 (Node 4) 6 y1 
otherwise
2 s.t. x25  x35  x56 (Node 5)

0
1 if operations research is taken
2 s.t. x46  x56  x0 (Node 6) 6 y2 
otherwise
2 s.t. x12  500, x13  400, x24  300,

0
1 if data structure is taken
2 s.t. x25  250, x34  200, x35  150, 6 y3 
otherwise
2 s.t. x46  400, x56  350

0
All variables  0 1 if business statistics is taken
6 y4 
otherwise
5 The MST consists of the following arcs: N.Y.–Clev.,
N.Y.–Nash., Nash.–Dal., Dal.–St.L., Dal.–Pho., Pho.–L.A.,
0
1 if computer simulation is taken
and S.L.C.–L.A. Total length of the MST is 4,300. 6 y5 
otherwise


1 if introduction to computer
Chapter 9 6 y6  programming is taken
0 otherwise
SECTION 9.2

0
1 if forecasting is taken
6 y7 

1 if player i starts otherwise
1 Let x1 
0 otherwise
Then the appropriate IP is
Then the appropriate IP is min z  y1  y2  y3  y4  y5  y6  y7
max z  3x1  2x2  2x3  x4  3x5  3x6  x7 s.t. y1  y2  y3  y4  y7  2 (Math)
s.t. x1  x3  x5  x7  4 (Guards) s.t. y2  y4  y5  y7  2 (OR)
s.t. x3  x4  x5  x6  x7  2 (Forwards) s.t. y3  y5  y6  2 (Computers)
s.t. x2  x4  x6  1 (Center) s.t. y4  y1
s.t. x1  x2  x3  x4  x5  x6  x7  5 s.t. y5  y6
s.t. 3x1  2x2  2x3  x4  3x5  3x6  3x7  10
(Ballhandling)

1386 APPENDIX 3 Answers to Selected Problems

 s.t. y3  y6 SECTION 9.5
s.t. y7  y4 2 max z  60x1  48x2  14x3  31x4  10x5
y1, y2, . . . , y7  0 or 1 s.t. 800x1  600x2  300x3  400x4  200x5  1,100
10 Add the constraints x  y  3  Mz, 2x  5y  xi  0 or 1
12  M(1  z), z  0 or 1, where M is a large positive
number. Optimal solution is z  79, x2  x4  1, x1  x3 
x5  0.
11 Add the constraints y  3  Mz, 3  x  (1  z)M,
z  0 or 1, where M is a large positive number.
SECTION 9.6
13 Let xi  number of workers employed on line i
1 Do jobs in the following order: job 2, job 1, job 3, and

1 if line i is used
yi  job 4. Total delay is 20 minutes.
0 otherwise
2 LFR—LFP—LP—LR—LFR has a total cost of $330.
Then the appropriate IP is
8 Warehouse 1 to factory 1, warehouse 2 to factory 3,
min z  1,000y1  2,000y2  500x1  900x2 warehouse 3 to factory 4, warehouse 4 to factory 2,
13 s.t. 20x1  50x2  120 warehouse 5 to factory 5 has a total cost of $35,000.
13 s.t. 30x1  35x2  150
SECTION 9.7
13 s.t. 40x1  45x2  200
1 z  4, x1  x2  x4  x5  1, x3  0.
13 s.t. x1  7y1
2 z  3, x1  x3  1, x2  0.
13 s.t. x2  7y2
x1, x2  0; y1, y2  0 or 1 SECTION 9.8


1 if disk i is used
14 a Let xi  1 z  110, x1  4, x2  3.
0 otherwise
Then the appropriate IP is REVIEW PROBLEMS
min z  3x1  5x2  x3  2x4  x5  4x6
0
1 if gymnast i enters both events
min z   3x7  x8  2x9  2x10 3 Let zi 
otherwise
s.t. x1  x2  x4  x5  x8  x9  1 (File 1)
0
1 if gymnast i enters only balance beam
3 xi 
s.t. x1  x3  1 (File 2) otherwise
s.t. x2  x5  x7  x10  1 (File 3)
0
1 if gymnast i enters only floor exercises
3 yi 
s.t. x3  x6  x8  1 (File 4) otherwise
s.t. x1  x2  x4  x6  x7  x9  x10  1 (File 5) Then the appropriate IP is
xi  0 or 1 (i  1, 2, . . . , 10) max z  16.7z1  17.7z2      17.7z6  8.8x1
b Add the constraints 1  x2  2y, x3  x5  max z   9.4x2      9.1x6  7.9y1
2(1  y), y  0 or 1 (need M  2, because x3  x5 
2 is possible). We could also have added the constraints max z   8.3y2      8.6y6
x2  x3 and x2  x5. s.t. z1  z2      z6  3
s.t. x1  x2      x6  1
SECTION 9.3
s.t. y1  y2      y6  1
1 z  20, x1  4, x2  0.
s.t. x1  y1  z1  1
2 z  $400,000. (x1  6, x2  1) and (x1  4, x2  2)
s.t. x2  y2  z2  1
are both optimal solutions.
s.t. x3  y3  z3  1
4 z  8, x1  2, x2  0.
s.t. x4  y4  z4  1
SECTION 9.4 s.t. x5  y5  z5  1
1 z  5.6, x1  1.2, x2  2. s.t. x6  y6  z6  1
All variables 0 or 1

APPENDIX 3 Answers to Selected Problems 1387

 26 Let


1 if students from district i
4 Let xij  are sent to school j
0 otherwise
1 if building i is started during year t
0 otherwise xit 
Then the appropriate IP is
Then the appropriate IP (in thousands) is
min z  110x11  220x12  37.5x21  127.5x22
max z  100x11  50x12  60x21  30x22  40x31
min z   80x31  80x32  117x41  36x42
s.t. 30x11  20x21  20x31  60 (Year 1 workers)
min z   135x51  54x52
s.t. 30(x11  x12)  20(x21  x22)
s.t. 110x11  75x21  100x31  90x41  90x51  150
(School 1  150 students) s.t.  20(x31  x32)  60 (Year 2 workers)
s.t. 110x12  75x22  100x32  90x42  90x52  150 s.t. 30(x12  x13)  20(x22  x23)
(School 2  150 students) s.t.  20(x31  x32  x33)  60 (Year 3 workers)
30x11  5x21  10x31  40x41  30x51 30(x13  x14)  20(x23  x24)
0.20   , s.t.
110x11  75x21  100x31  90x41  90x51
s.t.  20(x32  x33  x34)  60 (Year 4 workers)
or 0  8x11  10x21  10x31  22x41  12x51


x11  x21  x31  1
30x12  5x22  10x32  40x42  30x52 (No more than one building
0.20   , s.t. s.t. x12  x22  x32  1
110x12  75x22  100x32  90x42  90x52 begins during each year)
s.t. x13  x23  x33  1
or 0  8x12  10x22  10x32  22x42  12x52 s.t.
x11  x12  1


x11  x12  x13  x14  1
x21  x22  1 x21  x22  x23  x24  1 (Each building is
started at most once)
x31  x32  1 x31  x32  x33  x34  1
x41  x42  1 s.t. x21  x22  x23  1
(Building 2 is finished
by end of year 4)
x51  x52  1
All variables 0 or 1
All variables 0 or 1

0
1 if truck i is used
Let yi 

1 if RS is signed 27
5 Let x1  otherwise
0 otherwise

0
1 if truck i is used to deliver to grocer j
0
1 if BS is signed Let xij 
5 Let x2  27
otherwise
otherwise

0
1 if DE is signed Then the appropriate IP is
5 Let x3 
otherwise min z  45y1  50y2  55y3  60y4


1 if ST is signed s.t. 100x11  200x12  300x13  500x14  800x15
5 Let x4 
0 otherwise  400y1
100x21  200x22  300x23  500x24  800x25
0
1 if TS is signed s.t.
5 Let x5   500y2
otherwise
Then the appropriate IP is s.t. 100x31  200x32  300x33  500x34  800x35
 600y3
max z  6x1  5x2  3x3  3x4  2x5
s.t. 100x41  200x42  300x43  500x44  800x45
s.t. 6x1  4x2  3x3  2x4  2x5  12  1,100y4
s.t. 6x1  4x2  x3  x4  2x5  2 s.t. x11  x21  x31  x41  1
s.t. 6x1  x2  x3  2x4  2x5  2 s.t. x12  x22  x32  x42  1
s.t. 6x1  x2  3x3  2x4  2x5  1 s.t. x13  x23  x33  x43  1
All variables 0 or 1 s.t. x14  x24  x34  x44  1
10 Use two 20¢ coins, one 50¢ coin, and one 1¢ coin. s.t. x15  x25  x35  x45  1
13 Infeasible. All variables 0 or 1

1388 APPENDIX 3 Answers to Selected Problems

Chapter 10 Chapter 11
SECTIONS 10.1 AND 10.2 SECTION 11.1
1 z  11, x1  3, x2  0, x3  2. 1 3.
2 z  10, x1  2, x2  2. 3 a xex  ex.
3
e .
SECTION 10.3 x
x2 x3
2 Let xi  number of 15-ft boards cut according to 5 ln(1  x)  x    3 for some p between 1 and
1  x. 2 3p
combination i, where
3-Ft 5-Ft 8-Ft 7 a k.
Combination Boards Boards Boards b The maximum size of the market (as measured in
sales per year).
1 0 1 1
2 2 0 1 9 The machine time is the better buy.
3 0 3 0
4 1 2 0 SECTION 11.2
5 3 1 0 1 a Let S  soap opera ads and F  football ads. Then
6 5 0 0 we want to solve the following LP:
Then we want to solve min z  50S  100F
min z  x1  x2  x3  x4  x5  x6 s.t. 5S1/2  17F1/2  40 (Men)
s.t. 2x2  x4  3x5  5x6  10 s.t. 20S1/2  7F1/2  60 (Women)
s.t. x1  3x3  2x4  x5  20 S, F  0
s.t. x1  x2  15 b Since doubling S does not double the contribution of
S to the constraints, we are violating the Proportionality
s.t. xi  0 Assumption. Additivity is not violated.
The optimal solution is z  55

3 , x1  10, x2  5, x3  10
3 . 5 a  b  c  20.

SECTION 10.4 SECTION 11.3

1 z  40, x1  3, x2  2, x3  3. 1 Convex.
3 z  15, x1  0, x2  0, x3  3. 2 Neither convex nor concave.
5 Concave.
SECTION 10.5
8 Concave.
1 z  29.5, x1  2, x2  0.5, x3  4, x4  x5  0.
3 z 81

13 , x1  12

13 , x2  11

13. SECTION 11.4

1 If fixed cost is $5,000, spend $10,000 on advertising. If

SECTION 10.6 fixed cost is $20,000, don’t spend any money on advertising.
1 y1  x1  [38 1 3
4 8 ]. 2 Without tax, produce 12.25 units; with tax, produce 12
units.
REVIEW PROBLEMS 5 z  1, x  1.
1 z  9, x1  x3  0, x2  3.
SECTION 11.5
3 z  12, x2  2, x4  10, x1  x3  0.
2 After four iterations, the interval of uncertainty is
5 Maximum profit is $540. Optimal production levels are
[0.42, 0.17).
Product 1 at plant 1  5
3 units
Product 1 at plant 2  10

3
0
units
Product 2 at plant 1  29

9
0
units
Product 2 at plant 2  0 units

APPENDIX 3 Answers to Selected Problems 1389

SECTION 11.6 s.t. 11  12  13  14  15  1
x1  x2      xn s.t. 21  22  23  24  25  26  1
1 x   ,
n All variables nonnegative
y1  y2      yn Adjacency Assumption
1 y   .
n
3 q1  98.5, q2  1. SECTION 11.12
6 (0, 0) is a saddle point. (32, 32) and (32, 32) are each a 1 x  [0
1
1] and x2  [13 56].
local minimum.
SECTION 11.13
SECTION 11.7
2 First we attempt to maximize output by solving
3 Successive points are (12, 34), (34, 34), (34, 78).
max z  20x11  12x12  10x13  12x21  15x22
SECTION 11.8 max z   9x23  6x31  5x32  10x33
10 s.t. x11  x12  x13  1
2 LK ;
3
produce 10 machines.
s.t. x21  x22  x23  1
900 400 131/2
4 x1  , x2  ,     1. An extra dollar
13 13 2 s.t. x31  x32  x33  1
spent on promotion would increase profit by approximately
s.t. x11  x21  x31  1

131/2
$   1 .
2  s.t. x12  x22  x32  1

SECTION 11.9 s.t. x13  x23  x33  1

All xij  0
1 Capacity  27.5 kwh. Peak price  $65. Off-peak price
 $20. Here xij fraction of day worker i spends working on product
21/2 21/2 j. This LP has optimal solution x11  x22  x33  1, with
2 z  21/2, x1  , x2  . Output  45 and Happiness  6  5  8  19. Thus, point
2 2
(45, 19) is on a tradeoff curve. Now add constraint
6 z  12, x1  12, x2  32.
6x11  8x12  10x13  6x21  5x22  9x23  9x31
SECTION 11.10  10x32  8x33  HAPP

1 min z  0.09x21  0.04x22  0.01x32 where HAPP  20, 25, and 26. (HAPP cannot exceed 26.)
This yields four points on the tradeoff curve in Figure 11.
 0.012x1x2  0.008x1x3  0.010x2x3 Value of Output is optimal z-value for each LP.
s.t. x1  x2  x3  0
x1  x2  x3  100 FIGURE 11
x1, x2, x3  0 27

4 p1  $292.81, p2  $158.33. Pay no money for an 26

additional hour of labor. Pay up to (approximately) $53.81
for another chip. 25

SECTION 11.11 24
Happiness

1 Using grid points 0, 0.5, 1, 1.5, and 2 for x1 and grid 23

points 0, 0.5, 1, 1.5, 2, and 2.5 for x2, we obtain the following
approximating problem: 22

min z  0.2512  13  2.2514  415  0.2522 21

min z   23  2.2524  425  6.2526

20
s.t. 0.2512  13  2.2514  415  2(0.2522
s.t. 0.2512  23  2.2524  425  6.2526)  4 19
27 29 31 33 35 37 39 41 43 45

s.t. 0.2512  13  2.2514  415  0.2522 Output

s.t. 0.2512  23  2.2524  425  6.2526  6

1390 APPENDIX 3 Answers to Selected Problems

REVIEW PROBLEMS 1
7 a .
6
2 Locate the store at point 7. In general, locate store at b .004996.
arithmetic mean of the location of all customers.
3 a Use 983 units of raw material, sell 943 units of product Chapter 13
1, and sell 125 units of product 2.
c Pay slightly less than $5 for an extra unit of raw SECTION 13.1
material.
5 [1.18, 1.63). 1 Maximin decision; small campaign. Maximax decision:
large campaign. Minimax regret decision: large campaign.
7 Produce 20 units during each of the three months.
2 Maximin decision; don’t build. Maximax decision:
18 Locate the store at point 5; in general, locate store at build. Maximax regret decision: build. Expected value
the median of the customer’s locations. decision: build.
5 Maximin decision: $6,000, $8,000, or $11,000 bid.
Chapter 12 Maximax decision: $11,000 bid. Minimax regret decision:
$11,000 bid. Expected value decision: $11,000 bid.
SECTION 12.1
SECTION 13.2
e 1 10
1  . 1 a Risk-averse.
2
b Prefer L1; risk premium for L2  $339.
dh
3 Ih   . 2 a Risk-seeking.
2
b Prefer L2; risk premium for L2  $235.
SECTION 12.2 6 b $1,900.
7 Take statistics course.
1 2y(2y  y2)  3y  2(y2  y).
13 L1 is preferred.
SECTION 12.3
SECTION 13.4
1 a 29.
c No. 1 Hire the geologist. If she gives a favorable report, drill;
e 14. if she gives an unfavorable report, don’t drill. Expected
profit is $180,000; EVSI  $20,000; EVPI  $55,000.
SECTION 12.4 4 Market without testing. Expected profit  $16,000;
EVSI  $3,800; EVPI  $14,000.
2
1 .
3 9 Play daringly during the first game. If he wins the first
3 .001. game, play conservatively during the second game. If he
loses the first game, play daringly during the second game.
If tied after two games, play daringly during the third game.
SECTION 12.5
29 0
12 a Buy the gold now.
1 a Let S  number sold. Then E(S)  ,
3
and b Wait for Congress and (if possible) buy the gold
200
var S  9
.
later.
3 a F(a)  0 for a  0, F(a)  1  ea for a  0.
b E(X)  1 var X  1. SECTION 13.5
c e1  e2.
2 Hire the geologist. If he predicts an earthquake, build at
Roy Rogers; if he predicts no earthquake, build at Diablo.
SECTION 12.6 Expected total cost  $13,900,000. EVSI  $1,100,000;
EVPI  $2,000,000.
2 .8749.
4 Hire the firm. If it predicts a hit, air the show. If it
REVIEW PROBLEMS
predicts a flop, don’t air the show. Expected profit 
$35,000; EVSI  $50,000; EVPI  $75,000.
25
3 .
2
5 a E(X)  85; var X  9,000. SECTION 13.6


b P Z  
91  85
(9,000)1/2 
 .476. 1 c National’s utility function is of the form .3u1(x1)
 .5u2(x2)  .2u1(x1)u2(x2).

APPENDIX 3 Answers to Selected Problems 1391

6 d k3 0. e Gunner’s LP is
max z  v
SECTION 13.7
s.t. v  x1
1 a Professor 2 should receive the bigger raise. s.t. v  x1  x2
c Pairwise comparison matrix is consistent.
s.t. v  x2  x3
REVIEW PROBLEMS s.t. v  x3  x4
s.t. v  x4
1 a Invest in money market fund.
b Invest in gold. s.t. x1  x2  x3  x4  1
c Money market fund has a maximum regret of $500. s.t.x1, x2, x3, x4  0; v urs
d All investments have the same expected return. Soldier’s LP is
2 Because of the risk-averse nature of the utility function, min w
invest in the least risky investment (the money market fund).
s.t. w  y1  y2
5 a Invest all money in stocks.
b Indifferent between hiring and not hiring forecaster. s.t. w  y2  y3
Expected final asset position  $1,160,000; EVSI  s.t. w  y3  y4
$10,000; EVPI  $20,000.
s.t. w  y4  y5
12 If all potential litterers are risk-averse, then raising the
fine will result in the larger decrease in littering. s.t. y1  y2  y3  y4  y5  1
s.t.y1, y2, y3, y4, y5  0; w urs

Chapter 14 3 Value to row player  52. Row player’s optimal strategy

is (12, 12), and column player’s optimal strategy is (34, 14, 0).
SECTION 14.1
SECTION 14.4
1 Value to row player  2. Row player plays row 1, and
column player plays column 1. 1 (9, 1) is an equilibrium point.
2 Value to row player  6. Row player plays row 2, and 3 This is a Prisoner’s Dilemma game, with the equilibrium
column player plays column 1 or column 3. point occurring where each borough opposes the other
borough’s bond issues. Reward is $0 to each borough.
SECTION 14.2
SECTION 14.7
1 Value to row player  43. Row player’s optimal strategy
is (23, 13) and column player’s optimal strategy is (23, 13, 0). 2 Core consists of the point (25, 25, 25, 25).
8 State’s optimal strategy is, with probability 1
 ,
play A
2
3 a v({ })  $0; v({1})  v({2})  v({3})  $2;
first and B second; with probability 12, play B first and A v({1, 2})  v({2, 3})  v({1, 3})  $2; v({1, 2, 3})
second. Ivy’s optimal strategy is, with probability 12, play X  $3.
first and Y second; with probability 12, play X second and Y b and c The Shapley value gives $1 to each player.
first. Value of game to State  12. The core is ($1, $1, $1).
7 Assuming that the runway costs $1/ft, the Shapley value
SECTION 14.3 recommends the following fees per landing: type 1, $20;
240 440 1,140
type 2, $7
; type 3, $; type 4, $.
7 7
1 a
Soldier REVIEW PROBLEMS
Gunner 1 2 3 4 5
1 Both stores will be located at point B, and the two firms
Spot A 1 1 0 0 0 will each have 26 customers.
Spot B 0 1 1 0 0 3 b Value to row player  151. For each player, the
Spot C 0 0 1 1 0 optimal strategy is (141 , 141 , 131).
Spot D 0 0 0 1 1
6 a v({ })  v({49})  v({50})  v({1})  v({1, 49})
b Columns 2 and 4 are dominated.  0; v({1, 50})  v({49, 50})  v({1, 49, 50})  1.
c Expected value to the gunner  13. b Core consists of point (0, 0, 1).
d Always firing at A. c Shapley value gives 16 to player 1, 16 to player 2, and
2
 to player 3.
3

1392 APPENDIX 3 Answers to Selected Problems

Chapter 15 SECTION 16.4

SECTION 15.2 2 $775,000, using F(.25)  .60.

3 107.5 trees, using F(.25)  .60.
1 a 4,000 gallons.
b 12 orders per year.
SECTION 16.5
c One month.
1/2
e For a 2-week lead time, reorder point  48 ,000
 
26 1
2
Locate at .
1,846.15 gallons. For a 10-week lead time, reorder point 2
 1,230.77 gallons.
30 SECTION 16.6
3 a Send out   4.24 trucks per hour.
7.07
b Send out 30
  6 trucks per hour. 1 Order quantity  45.61, reorder point  32.6, safety
5
stock  12.6.
12 b Six trainees in each program.
2 a Reorder print  57.9, safety stock  17.9.
c Run 4.5 programs per year.
b Reorder print  60.60, safety stock  20.60.
d 2.25 trainees.

SECTION 15.3 SECTION 16.7

1 Order 300 boxes per year. Place 3.2 orders per year. 1
5 Order 100 thermometers. Reorder
SLM1 Point
SECTION 15.4 80% 11.06
2 Optimal run size  692.82. Do 34.64 runs per year. 90% 16.46
95% 20.24
SECTION 15.5 99% 26.30

2 Whenever dealer is 10 cars short, an order for 50 cars
should be placed. Maximum shortage will be 10 cars.
SECTION 15.6
1 Demand is too lumpy to justify using EOQ.
For SLM2  0.5 stockout per year, the reorder point is 32.06.
3 a SLM1  98.75%.
b r  20 is the smallest reorder point with SLM1
exceeding 95%.
c Need a reorder point of at least 30 units.

REVIEW PROBLEMS SECTION 16.8

1 a 268.33 desks. 1 R  0.23 years and S  3,241.

b 22.36 orders per year.
c 2(22.36)(300)  $13,416. SECTION 16.9
d For 1-week lead time, reorder point  115.38 desks;
1 Type A items: 1 and 2; type B items: 3 through 6; type
for 5-week lead time, reorder point  40.93 desks.
C items: 7 through 10. See Figure 12.
e Order 342.05 desks 17.54 times per year.
3 The EOQ for the lowest price is optimal. Thus, 417.79 FIGURE 12
cameras should be ordered.

Chapter 16
SECTION 16.2

1 a q  6.
b q  2.
c E(q) is not a convex function of q.

SECTION 16.3

1 Order 35 cars.
3 q* will decrease.
5 a Order 60 cells.

APPENDIX 3 Answers to Selected Problems 1393

SECTION 16.10 SECTION 17.4

1 a See Figure 13. 1 2.

2 Yes.
FIGURE 13 3 a State 4.
b States 1, 2, 3, 5, and 6.
c {1, 3, 5} and {2, 6}.
4 P1 is ergodic; P2 is not ergodic.

SECTION 17.5
38 90 55
1 Urban, ; suburban, ; rural, .
183 183 183
3 a State 1, 35; state 2, 2
.
5
4 Replace a fair car.
7 Expected price of stock 1  $16.67; expected price of
stock 2  $16.00.
REVIEW PROBLEMS
SECTION 17.6
1 a Bake 40 dozen cookies.
1 a 1.11  0.99  0.99  0.88  3.97.
b Bake 58.6 dozen cookies.
b .748.
c Bake 55 dozen cookies.
2 1  0.80  18  19.80 years.
3 a Order quantity  707.11, shortage cost  $12.41.
b Order quantity  707.11. Assuming a penalty for 11 a 14.3%.
lost sales of 8  5  $3, shortage cost  $9.12. b Reducing warranty period will save $715,000 
c Reorder point of zero will do the job. $392,500  $322,500.

SECTION 17.7
Chapter 17
1 7,778 freshmen, 7,469 sophomores, 8,057 juniors, and
SECTION 17.2 7,162 seniors.
2 Each working adult must contribute $2,000 more.
Sunny Cloudy

1
Sunny
Cloudy  .90
.20
.10
.80  REVIEW PROBLEMS
260
1 An average of  tools per day will be produced.
3
2 State
2 a .815.
0 1 2 3 4 b Annual profit without warranty  (3,000)(total

 
0 0 0 13 13 13 market size)(14). Annual profit with warranty 
1 0 0 13 13 13 (2,700)(total market size)(49). Profit with warranty is
2 13 13 13 0 0 larger.
3 0 13 13 13 0 3 Value of star  $4,400,000; value of starter 
4 0 0 13 13 13 $3,199,000; value of substitute  $1,600,000.
4 Let SC denote that yesterday was sunny and today is 4 a Each year, 1,638,270 new books, 1,474,443 once-
cloudy, and so on. used books, 1,179,554 twice-used books, and 707,733
SS SC CS CC thrice-used books will be sold.

 
SS .95 .05 0 0
SC 0 0 .40 .60 Chapter 18
CS .70 .30 0 0
CC 0 0 .20 .80 SECTION 18.1

SECTION 17.3 1 Begin by picking up 4 matches. On each successive

turn, pick up 5  (number of matches picked up by opponent
1 a Urban, .651; suburban, .258; rural, .091. on last turn).
b 31.5%. 2 The players began with $16.25, $8.75, and $5.00.

1394 APPENDIX 3 Answers to Selected Problems

SECTION 18.2 SECTION 18.6
1 1–3–5–8–10, 1–4–6–9–10, and 1–4–5–8–10 are all 1 Let ft(i) be the maximum expected net profit earned
shortest paths from node 1 to node 10 (each has length 11). during years t, t  1, . . . , 10, given that Sunco has i barrels
The path 3–5–8–10 is the shortest path from node 3 to node of reserves at the beginning of year t. Then
10 (this path has length 7).
f10(i)  max {xp10  c(x)}
x
3 Bloomington–Indianapolis–Dayton–Toledo–Cleveland
takes 8 hours. where x must satisfy 0  x  i. For t  9,
ft(i)  max {xpt  c(x)  ft1(i  bt  x)} (1)
x
SECTION 18.3 where 0  x  i. We use Equation (1) to work backward
until f1(i0) is determined. If discounting is allowed, let b 
1 Produce no units during month 1, 3 units during month the discount factor. Then we redefine ft (i) to be measured in
2, no units during month 3, and 4 units during month 4. terms of year t dollars. Then we replace (1) with (1
):
Total cost is $15.00
ft(i)  max {xpt  c(x)  bft1(i  bt  x)} (1)
2 Month 1, 200 radios; month 2, 600 radios; month 3, no x
radios. Total cost is $8,950. where 0  x  b.
3 a Produce 1 unit. Cost associated with arc is $4.50. 2 b Let ft(d) be the maximum utility that can be earned
during years t, t  1, . . . , 10, given that d dollars are
available at the beginning of year t (including year t
SECTION 18.4
income). During year 10, it makes sense to consume all
1 Site 1, $1 million; site 2, $2 million; site 3, $1 million. available money (after all, there is no future). Thus,
Total revenue is $24 million. f10(d)  ln d. For t  9,

2 Obtain a benefit of 10 with two type 1 items or two type ft(d)  max {ln c  ft1(1.1(d  c)  i)}
c
2 and one type 3 item. where 0  c  d. We work backward from the f10()’s
3 a See Figure 14. to f1(D).
5 French, 1 hour; English, no hours; statistics, 3 hours.
There is a .711 chance of passing at least one course.
FIGURE 14
7 Define ft (w) to be the maximum net profit (revenues less
costs) obtained from the steer during weeks t, t  1, . . . ,
10, given that the steer weighs w pounds at the beginning of
week t. Now
f10(w)  max {10g(w, p)  c(p)}
p
where 0  p. Then for t  9,
ft(w)  max {c(p)  ft1(g(w, p))}
p
Farmer Jones should work backward until f1(w0) has been
computed.
8 Define ft(i, d) to be the maximum number of loyal
customers at the end of month 12, given that there are i loyal
customers at the beginning of month t and d dollars available
to spend on advertising during months t, t  1, . . . , 12. If
there is only one month left, all available funds should be
spent during that month. This yields
f12(i, d)  (1  p(d))i  (N  i)q(d)
For t  11,
ft(i, d)  max {ft1[(1  p(x))i  (N  i)q(x), d  x]}
x
where 0  x  d. We work backward until f1(i0, D) has been
determined.
5 One type 2 item and one type 3 item yield a benefit 10 Let ft(it, xt1) be the minimum cost incurred during
of 75. months t, t  1, . . . , 12, given that inventory at the
beginning of month t is it, and production during month
SECTION 18.5 t  1 was xt1. Then
f12(i12, x11)  min {c12x12  5|x12  x11|
2 Trade in car whenever it is two years old (at times 2, 4, x12
and 6). Net cost is $14,400.  h12(i12  x12  d12)}

APPENDIX 3 Answers to Selected Problems 1395

where x12 must satisfy x12  0 and i  x12  d12. For 2 At any time, produce the number of units needed to
t  11, bring the period’s stock level (before the period’s demand is
met) to 2 units.
ft(it, xt1)
 min {ctxt  5|xt  xt1|  ht(i  xt  dt) SECTION 19.3
xt
 ft1(it  xt  dt, xt)}
1 Ulanowsky should play boldly during the first game. If
where xt must satisfy xt  0 and it  xt  dt. We work he wins the first game, then he should play conservatively
backward until f1(20, 20) has been computed. in the second game. If he loses the first game, then he should
play boldly in the second game. If there is a tie-breaking
SECTION 18.7 game, Ulanowsky should play boldly. His chance of winning
the match is .537.
1 If initial inventory is 200 units, only modification is to
2 On the first toss, Dickie should bet $2. If he loses, he is
produce 200 fewer units during period 1.
wiped out, but if he wins, he bets $1 on the second toss. If
2 The Wagner–Whitin and Silver–Meal methods both he wins on the second toss, he stops. If he loses on the
yield the following production schedule: period 1, 90 units; second toss, he bets $2 on the third toss.
period 4, 230 units. Total cost is $176.
SECTION 19.4
REVIEW PROBLEMS
2 Let ft(d) be the maximum expected asset position of the
1 Shortest path from node 1 to node 10 is 1–4–8–10. firm at the end of year 10, given that at the beginning of
Shortest path from node 2 to node 10 is 2–5–8–10. year t, the firm has d dollars in assets. Then
2 Month 2, 1 unit; month 3, 4 units. Total cost is $12. f10(d)  max p
i 
y qy(d  i  y)
 (1  p) qy(d  i  y)
4 For 6 flights, the airline earns $540 with 3 Miami, 2
L.A. and 1 N.Y. flight; or 3 Miami and 3 L.A. flights. For y
four flights, the airline earns $375 with 2 Miami and 2 L.A.
where 0  i  d. For t  9,
flights.
5 Without the 20¢ piece, use one 50¢, one 25¢, one 10¢,
ft(d)  max p
i 
y qy ft1(d  i  y)
 (1  p) qy ft1(d  i  y)
one 5¢, and one 1¢ piece. With the 20¢ piece, use one 50¢,
two 20¢, and one 1¢ piece. y
7 a Let ft(w) be the minimum cost incurred in meeting We work backward until f1(10,000) has been computed.
demands for the years t, t  1, . . . , 5, given that (before
5 We should always do maintenance on a running machine
hiring and firing for year t) w workers are available.
and always repair a broken machine.
ht  workers hired at beginning of year t
6 Let ft(p) be the maximum expected revenue earned from
dt  workers fired at beginning of year t selling a share of Wivco stock during days t, t  1, . . . , 30,
given that the price of a share at the beginning of day t is p
wt  workers required during year t
dollars. Then
Then
f30(p)  p (Sell stock)
ft (w)  min {10,000ht  20,000dt
ht ,dt and for t  29,
 30,000(w  ht  dt)  ft1(ht  .9(w  dt))

 
p (Sell stock)
where ht and dt must satisfy 0  ht, 0  dt  w, and ft(p)  max x q(x)ft1((1  x/100)p) (Keep stock)
w  ht  dt  wt.
We work backward until f1(10) has been determined, and we
continue until the optimal action is to sell the stock.
Chapter 19 7 Sara should not accept the first cat, but any later cat that
is the best Sara has seen so far should be accepted. The
SECTION 19.1 probability that Sara will get her preferred cat is 2141 .

1 Two gallons of milk to each store. 8 Define ft (i) to be the minimum net expected cost incurred
during periods t, t  1, . . . , 100, given that the inventory
2 $2 million to investment 1, $0 to investment 2, and $2 level is i at the beginning of period t. Then

million to investment 3.
f100(i)  min
x
qd(i  x  d  rd)
dix
SECTION 19.2  qd(p(d  i  x)  r(i  x))  c(x)
d ix
1 Produce 3 units during period 1, no units during period
2, and 1 unit during period 3.

1396 APPENDIX 3 Answers to Selected Problems

where x  0. For t  99, Chapter 20
x  di+x
ft(i)  min qd(i  x  d  rd)
SECTION 20.2
 (p(d  i  x)  r(i  x))qd  c(x)
d ix 6
1 .
 qdft1(i  x  d) 7
dix

 qdft1(0) 2
555
 minutes.
d ix 11
where x  0. We work backward until f1(0) has been 2e2
determined. 4 a   .09.
3
10 Define ft(d) to be the maximum expected number of b 1  e2  2e2  .594.
units sold in markets t, t  1, . . . , T, given that d dollars
are available to spend on these markets. Then
SECTION 20.3
fT(d)  cTkTpT(d)
121
and for t  T  1, 2 b .
144
ft(d)  max {ctkt pt (x)  ft1(d  x)} 1
x
c .
where 0  x  d. We work backward until f1(D) has been 144
computed.
SECTION 20.4
SECTION 19.5
5
1 a .
1 a 6
25
Period’s Period’s b  passengers.
6
Beginning Production
1
Inventory Level c  minute.
2
0 4 3 a Unchanged.
1 3 b Cut in half.
2 0 c Unchanged.
3 0 8 Two checkpoints.
2 a Always charge 11% interest rate on loans. 10 b 1 taxi.
c $120 per hour.
REVIEW PROBLEMS

1 Assign 2 sales reps to district 1, 1 sales rep to district 2, SECTION 20.5

and 2 sales reps to district 3. 1 a 1.75 customers per hour.
2 a Produce 2 units during period 1. During period 2, 114
b  .
130
produce 1 or 2 units if beginning inventory is 0; if beginning
inventory for period 2 is 1 or 2 units, produce no units 6 Barber 1’s average hourly revenue  $53.23. Barber 2’s
during period 2. During period 3, produce 1 unit if beginning average hourly revenue  $40.00.
inventory is 0; otherwise, produce no units during period 3.
SECTION 20.6
4 Let ft(b) be the maximum discounted net benefit earned
during years t, t  1, . . . , 2039, given that b barrels of oil 1 Two registers.
are available at the beginning of year t. Then
3 a Finance, 15 day; Marketing, 1
 day.
ft(b)  max {u(x)  d  bp(d) ft1(b  x 10
d, x b 0.078 day.
 500,000)  b(1  p(d)) ft1(b  x)} d .07.
where 0  d and 0  x  b. We work backward until 4 4 servers.
f2004(B) is determined and then compute the optimal 3
consumption strategy. 13  mechanic.
4

5 a Try to answer the first two questions and then stop. 14 System 1 is more efficient.
The expected amount of money won is $9,000.
SECTION 20.7

1 5,200 members.

APPENDIX 3 Answers to Selected Problems 1397

3 200 firms. The probability that there are at least 3,200 REVIEW PROBLEMS
firms is zero.
2 a 8 minutes.
(1.25)2e1.25
SECTION 20.8 b   .22.
2!
2
1  car.
3 c e0.75  .47.
5
4 a  customer.
6 3 a 3 windows.
b 9 minutes. b 3 windows.
1
c .
3
6 2 copiers
2
SECTION 20.9 7 a car.
3
b 8.2 minutes.
1 Superworker is better.
81
8 Rent the vacant lot. Then expected daily lost profit is
2 a .
125 21(20)(0.008)  $3.36.
6 1
b  puppies. 9  hour.
5 2
11 Have one crew of 100 workers.
SECTION 20.10
16 a Both lines are free 183 of time, one line is free 143 of
2 a 2.73 cars. the time, and both lines are busy 113 of the time.
b 0.06 hour. b 163 line.
4 11
 students. c 30(113 )  30

13
callers per hour.
3

SECTION 20.11
Chapter 21
1 15 fire engines.
7 b 2 copies. SECTION 21.4

1 Approximately 0.76 minute. The answer may vary

SECTION 20.12 according to the random numbers used in the computations.

1 Using four categories (each having ei  6), we accept 3 See Figure 15. Variables used in the model:
the hypothesis that the length of a telephone call is
1,024
VALUE  face value from the roll of the dice
exponential with mean   seconds.
24
WINS  total number of wins up to current
simulation
SECTION 20.15
LOSSES  total number of losses up to current
1 Tests spend an average of 189 hours in the system, research simulation
papers spend an average of 247 hours in the system, and class POINT  face value from the first roll
handouts spend an average of 223 hours in the system.
PWON  proportion of wins
2 Highest priority to k  1 customers, next highest priority
to k  2 customers, and so on.

1398 APPENDIX 3 Answers to Selected Problems

FIGURE 15

SECTION 21.5 ARM: Generate two random numbers, r1 and r2.

1 ITM: Generate a random number r. Set x*  3r1

If (r  0.5) then If (x*  1) then

Ifx  2r IfAccept x* as the random variate

else else


Ifx  3  2
2  2r IfIf (r2  32(1  r1)) then

endif IfIfAccept x* as the random variate

APPENDIX 3 Answers to Selected Problems 1399

 Ifelse FIGURE 16
IfIfReject x* and repeat the process
Ifendif
endif
2 Generate a random number r.
If (r  0.25) then
Ifx  2  4
r
else

Ifx  10  4
3  3r
endif
4
E (x) var x

After 250 variates 5.274 2.823

After 500 variates 5.318 2.755
After 1,000 variates 5.364 2.856
After 5,000 variates 5.344 2.887
Theoretical values 5.333 2.889

REVIEW PROBLEMS

1 0.70, 0.33, 0.04, 0.11, 0.30, 0.53, 0.44, 0.91, 0.90, 0.73.
3 See Figure 16. Variables used in the model:
DAY  current day in the simulation
EXC  total number of days with net demand
greater than 100
NRES  number of rooms reserved for the current
day in simulation
NNSH  number of no-shows for the current day in
simulation
NUSD  number of rooms used on the current day
in simulation
Chapter 24
PERCENT  percentage of days when net demand
exeeded 100
SECTION 24.4

1 a 4,000 cases.
b 3,980 cases.
2 a New base  452; new trend  43.6; new summer
seasonality  1.37.
b Forecast for winter quarter  377.44.
4 95% sure that December sales will be between 172.5
and 297.5.
12 a 106.
b 115.26.

SECTION 24.5

1 For each professor’s payday, compute

Actual customers

Forecast customers

1400 APPENDIX 3 Answers to Selected Problems

Average these ratios. Suppose we obtain 1.3. Then to obtain 3 a SALES  52,900  912.5T  9,859Q1  8,467Q2
forecast for a day on which professors are paid, compute a  20,129Q4, where T  quarter number and Qi 
forecast by our basic method and multiply this forecast by dummy variable for quarter i.
1.3. d The part (b) model has smaller standard error than
the part (a) model. Thus, the part (b) model will yield a
SECTION 24.8 better forecast.
4 SALES  e7 PRICE0.67.
1 a Estimate beta  0.88.
b Yes. 5 b Indicates positive autocorrelation.
c 45%.
e 16.1%.

APPENDIX 3 Answers to Selected Problems 1401