8

Linear Programming:
Sensitivity Analysis and Interpretation of Solution
MULTIPLE CHOICE
1. To solve a linear programming problem with thousands of variables and constraints
a. a personal computer can be used.
b. a mainframe computer is required.
c. the problem must be partitioned into subparts.
d. unique software would need to be developed.
ANSWE! a
T"#$%! %omputer solution
&. A negative dual price for a constraint in a minimi'ation problem means
a. as the right(hand side increases) the ob*ective function value will increase.
b. as the right(hand side decreases) the ob*ective function value will increase.
c. as the right(hand side increases) the ob*ective function value will decrease.
d. as the right(hand side decreases) the ob*ective function value will decrease.
ANSWE! a
T"#$%! +ual price
,. $f a decision variable is not positive in the optimal solution) its reduced cost is
a. what its ob*ective function value would need to be before it could become positive.
b. the amount its ob*ective function value would need to improve before it could become positive.
c. 'ero.
d. its dual price.
ANSWE! b
T"#$%! educed cost
-. A constraint with a positive slac. value
a. will have a positive dual price.
b. will have a negative dual price.
c. will have a dual price of 'ero.
d. has no restrictions for its dual price.
ANSWE! c
T"#$%! Slac. and dual price
1
2 Chapter 8 LP Sensitivity Analysis and Interpretation of Solution
/. The amount b0 which an ob*ective function coefficient can change before a different set of values for the
decision variables becomes optimal is the
a. optimal solution.
b. dual solution.
c. range of optimalit0.
d. range of feasibilit0.
ANSWE! c
T"#$%! ange of optimalit0
1. The range of feasibilit0 measures
a. the right(hand(side values for which the ob*ective function value will not change.
b. the right(hand(side values for which the values of the decision variables will not change.
c. the right(hand(side values for which the dual prices will not change.
d. each of the above is true.
ANSWE! c
T"#$%! ange of feasibilit0
2. The 1334 ule compares
a. proposed changes to allowed changes.
b. new values to original values.
c. ob*ective function changes to right(hand side changes.
d. dual prices to reduced costs.
ANSWE! a
T"#$%! Simultaneous changes
8. An ob*ective function reflects the relevant cost of labor hours used in production rather than treating them
as a sun. cost. The correct interpretation of the dual price associated with the labor hours constraint is
a. the ma5imum premium 6sa0 for overtime7 over the normal price that the compan0 would be
willing to pa0.
b. the upper limit on the total hourl0 wage the compan0 would pa0.
c. the reduction in hours that could be sustained before the solution would change.
d. the number of hours b0 which the right(hand side can change before there is a change in the
solution point.
ANSWE! a
T"#$%! +ual price
8. A section of output from The 9anagement Scientist is shown here.
:ariable ;ower ;imit %urrent :alue <pper ;imit
1 13 133 1&3
What will happen to the solution if the ob*ective function coefficient for variable 1 decreases b0 &3=
a. Nothing. The values of the decision variables) the dual prices) and the ob*ective function will all
remain the same.
b. The value of the ob*ective function will change) but the values of the decision variables and the
dual prices will remain the same.
c. The same decision variables will be positive) but their values) the ob*ective function value) and
the dual prices will change.
d. The problem will need to be resolved to find the new optimal solution and dual price.
ANSWE! b
T"#$%! ange of optimalit0
13. A section of output from The 9anagement Scientist is shown here.
%onstrain
t
;ower ;imit %urrent :alue <pper ;imit
& &-3 ,33 -&3
Chapter 8 LP Sensitivity Analysis and Interpretation of Solution 3
What will happen if the right(hand(side for constraint & increases b0 &33=
a. Nothing. The values of the decision variables) the dual prices) and the ob*ective function will all
remain the same.
b. The value of the ob*ective function will change) but the values of the decision variables and the
dual prices will remain the same.
c. The same decision variables will be positive) but their values) the ob*ective function value) and
the dual prices will change.
d. The problem will need to be resolved to find the new optimal solution and dual price.
ANSWE! d
T"#$%! ange of feasibilit0
11. The amount that the ob*ective function coefficient of a decision variable would have to improve before that
variable would have a positive value in the solution is the
a. dual price.
b. surplus variable.
c. reduced cost.
d. upper limit.
ANSWE! c
T"#$%! $nterpretation of computer output
1&. The dual price measures) per unit increase in the right hand side)
a. the increase in the value of the optimal solution.
b. the decrease in the value of the optimal solution.
c. the improvement in the value of the optimal solution.
d. the change in the value of the optimal solution.
ANSWE! c
T"#$%! $nterpretation of computer output
1,. Sensitivit0 anal0sis information in computer output is based on the assumption of
a. no coefficient change.
b. one coefficient change.
c. two coefficient change.
d. all coefficients change.
ANSWE! b
T"#$%! Simultaneous changes
1-. When the cost of a resource is sun.) then the dual price can be interpreted as the
a. minimum amount the firm should be willing to pa0 for one additional unit of the resource.
b. ma5imum amount the firm should be willing to pa0 for one additional unit of the resource.
c. minimum amount the firm should be willing to pa0 for multiple additional units of the resource.
d. ma5imum amount the firm should be willing to pa0 for multiple additional units of the resource.
ANSWE! b
T"#$%! +ual price
1/. The amount b0 which an ob*ective function coefficient would have to improve before it would be possible
for the corresponding variable to assume a positive value in the optimal solution is called the
a. reduced cost.
b. relevant cost.
c. sun. cost.
d. dual price.
ANSWE! a
T"#$%! educed cost
4 Chapter 8 LP Sensitivity Analysis and Interpretation of Solution
11. Which of the following is not a question answered b0 sensitivit0 anal0sis=
a. $f the right(hand side value of a constraint changes) will the ob*ective function value change=
b. "ver what range can a constraint>s right(hand side value without the constraint>s dual price
possibl0 changing=
c. ?0 how much will the ob*ective function value change if the right(hand side value of a constraint
changes be0ond the range of feasibilit0=
d. ?0 how much will the ob*ective function value change if a decision variable>s coefficient in the
ob*ective function changes within the range of optimalit0=
ANSWE! c
T"#$%! $nterpretation of computer output
TRUE!ALSE
1. "utput from a computer pac.age is precise and answers should never be rounded.
ANSWE! @alse
T"#$%! %omputer solution
&. The reduced cost for a positive decision variable is 3.
ANSWE! True
T"#$%! educed cost
,. When the right(hand sides of two constraints are each increased b0 one unit) the ob*ective function value
will be ad*usted b0 the sum of the constraints> dual prices.
ANSWE! @alse
T"#$%! Simultaneous changes
-. $f the range of feasibilit0 indicates that the original amount of a resource) which was &3) can increase b0
/) then the amount of the resource can increase to &/.
ANSWE! True
T"#$%! ange of feasibilit0
/. The 1334 ule does not impl0 that the optimal solution will necessaril0 change if the percentage e5ceeds
1334.
ANSWE! True
T"#$%! Simultaneous changes
1. @or an0 constraint) either its slac.Asurplus value must be 'ero or its dual price must be 'ero.
ANSWE! True
T"#$%! +ual price
2. A negative dual price indicates that increasing the right(hand side of the associated constraint would be
detrimental to the ob*ective.
ANSWE! True
T"#$%! +ual price
8. +ecision variables must be clearl0 defined before constraints can be written.
ANSWE! True
T"#$%! 9odel formulation
8. +ecreasing the ob*ective function coefficient of a variable to its lower limit will create a revised problem
that is unbounded.
ANSWE! @alse
T"#$%! ange of optimalit0
Chapter 8 LP Sensitivity Analysis and Interpretation of Solution 5
13. The dual price for a percentage constraint provides a direct answer to questions about the effect of
increases or decreases in that percentage.
ANSWE! @alse
T"#$%! +ual price
11. The dual price associated with a constraint is the improvement in the value of the solution per unit
decrease in the right(hand side of the constraint.
ANSWE! @alse
T"#$%! $nterpretation of computer output
1&. @or a minimi'ation problem) a positive dual price indicates the value of the ob*ective function will
increase.
ANSWE! @alse
T"#$%! $nterpretation of computer output((a second e5ample
1,. There is a dual price for ever0 decision variable in a model.
ANSWE! @alse
T"#$%! $nterpretation of computer output
1-. The amount of a sun. cost will var0 depending on the values of the decision variables.
ANSWE! @alse
T"#$%! %autionar0 note on the interpretation of dual prices
1/. $f the optimal value of a decision variable is 'ero and its reduced cost is 'ero) this indicates that
alternative optimal solutions e5ist.
ANSWE! True
T"#$%! $nterpretation of computer output
11. An0 change to the ob*ective function coefficient of a variable that is positive in the optimal solution will
change the optimal solution.
ANSWE! @alse
T"#$%! ange of optimalit0
12. elevant costs should be reflected in the ob*ective function) but sun. costs should not.
ANSWE! True
T"#$%! %autionar0 note on the interpretation of dual prices
18. $f the range of feasibilit0 for b1 is between 11 and ,2) then if b1 B && the optimal solution will not change
from the original optimal solution.
ANSWE! @alse
T"#$%! ight(hand sides
18. The 133 percent rule can be applied to changes in both ob*ective function coefficients and right(hand sides
at the same time.
ANSWE! @alse
T"#$%! Simultaneous changes
&3. $f the dual price for the right(hand side of a C constraint is 'ero) there is no upper limit on its range of
feasibilit0.
ANSWE! True
T"#$%! ight(hand sides
SHORT A"S#ER
6 Chapter 8 LP Sensitivity Analysis and Interpretation of Solution
1. +escribe each of the sections of output that come from The 9anagement Scientist and how 0ou would use
each.
T"#$%! $nterpretation of computer output
&. E5plain the connection between reduced costs and the range of optimalit0) and between dual prices and
the range of feasibilit0.
T"#$%! $nterpretation of computer output
,. E5plain the two interpretations of dual prices based on the accounting assumptions made in calculating
the ob*ective function coefficients.
T"#$%! +ual price
-. Dow can the interpretation of dual prices help provide an economic *ustification for new technolog0=
T"#$%! +ual price
/. Dow is sensitivit0 anal0sis used in linear programming= Eiven an e5ample of what t0pe of questions that
can be answered.
T"#$%! Sensitivit0 anal0sis
1. Dow would sensitivit0 anal0sis of a linear program be underta.en if one wishes to consider simultaneous
changes for both the right(hand(side values and ob*ective function.
T"#$%! Simultaneous sensitivit0 anal0sis
PRO$LEMS
1. $n a linear programming problem) the binding constraints for the optimal solution are
/F G ,H C ,3
&F G /H C &3
a. @ill in the blan.s in the following sentence!
As long as the slope of the ob*ective function sta0s between IIIIIII and IIIIIII) the current
optimal solution point will remain optimal.
b. Which of these ob*ective functions will lead to the same optimal solution=
17 &F G 1H &7 2F G 8H ,7 83F G 13H -7 &/F G ,/H
T"#$%! Eraphical sensitivit0 anal0sis
&. The optimal solution of the linear programming problem is at the intersection of constraints 1 and &.
9a5 &51 G 5&
s.t. -51 G 15& C -33
-51 G ,5& C 133
151 G &5& ,33
51 ) 5& J 3
a. "ver what range can the coefficient of 51 var0 before the current solution is no longer optimal=
b. "ver what range can the coefficient of 5& var0 before the current solution is no longer optimal=
c. %ompute the dual prices for the three constraints.
T"#$%! Eraphical sensitivit0 anal0sis
,. The binding constraints for this problem are the first and second.
Chapter 8 LP Sensitivity Analysis and Interpretation of Solution 7

9in 51 G &5&
s.t. 51 G 5& ,33
&51 G 5& -33
&51 G /5& C 2/3
51 ) 5& J 3

a. Keeping c& fi5ed at &) over what range can c1 var0 before there is a change in the optimal solution
point=
b. Keeping c1 fi5ed at 1) over what range can c& var0 before there is a change in the optimal solution
point=
c. $f the ob*ective function becomes 9in 1./51 G &5&) what will be the optimal values of 51) 5&) and
the ob*ective function=
d. $f the ob*ective function becomes 9in 251 G 15&) what constraints will be binding=
e. @ind the dual price for each constraint in the original problem.
T"#$%! Eraphical sensitivit0 anal0sis
-. E5cel>s Solver tool has been used in the spreadsheet below to solve a linear programming problem with a
ma5imi'ation ob*ective function and all C constraints.
Input Section
Objective Function Coefficients
F H
- 1
Constraints Avail.
L1 , / 13
L& , & -8
L, 1 1 &3
Output Section
ariables 1,.,,,,,, -
!rofit /,.,,,,,, &- 22.,,,,,,
Constraint "sa#e Slac$
L1 13 1.288E(11
L& -8 (&.18&E(11
L, 12.,,,,,, &.1111112
a. Eive the original linear programming problem.
b. Eive the complete optimal solution.
T"#$%! Spreadsheet solution of ;#s
% Chapter 8 LP Sensitivity Analysis and Interpretation of Solution
/. E5cel>s Solver tool has been used in the spreadsheet below to solve a linear programming problem with a
minimi'ation ob*ective function and all J constraints.
Input Section
Objective Function Coefficients
F H
/ -
Constraints &e'().
L1 - , 13
L& & / /3
L, 8 8 1--
Output Section
ariables 8.1 2.&
!rofit -8 &8.8 21.8
Constraint "sa#e Slac$
L1 13 1.,/E(11
L& //.& (/.&
L, 1-- (&.1&E(11
a. Eive the original linear programming problem.
b. Eive the complete optimal solution.
T"#$%! Spreadsheet solution of ;#s
1. <se the spreadsheet and Solver sensitivit0 report to answer these questions.
a. What is the cell formula for ?1&=
b. What is the cell formula for %1&=
c. What is the cell formula for +1&=
d. What is the cell formula for ?1/=
e. What is the cell formula for ?11=
f. What is the cell formula for ?12=
g. What is the optimal value for 51=
h. What is the optimal value for 5&=
i. Would 0ou pa0 M./3 each for up to 13 more units of resource 1=
*. $s it possible to figure the new ob*ective function value if the profit on product 1 increases b0 a
dollar) or do 0ou have to rerun Solver=
Chapter 8 LP Sensitivity Analysis and Interpretation of Solution *
Input Infor+ation
:ar. 1 :ar. & 6t0pe7 Avail.
%onstraint 1 & / C -3
%onstraint & , 1 C ,3
%onstraint , 1 1 J 1&
#rofit / -
Output Infor+ation
:ariables
#rofit B Total
esources <sed Sl.ASurp
%onstraint 1
%onstraint &
%onstraint ,
,icrosoft -.cel 7./ Sensitivit0 &eport
1or$s2eet3 4.3s7base..ls5S2eet1
Changing Cells
Final &e)uce) Objective Allo6able Allo6able
Cell 7a+e alue Cost Coefficient Increase 8ecrease
M?M1& :ariables :ariable 1 8.-11/,8-1& 3 / 2 ,.-
M%M1& :ariables :ariable & -.11/,8-11/ 3 - 8./ &.,,,,,,,,,
Constraints
Final S2a)o6 Constraint Allo6able Allo6able
Cell 7a+e alue !rice &.9. Si)e Increase 8ecrease
M?M1/ constraint 1 <sed -3 3./,8-11/,8 -3 113 2
M?M11 constraint & <sed ,3 1.,3218&,38 ,3 ,3 -.111111112
M?M12 constraint , <sed 1,.3218&,38 3 1& 1.3218&,322 1EG,3
T"#$%! Spreadsheet solution of ;#s
2. <se the following 9anagement Scientist output to answer the questions.
;$NEA #"EA99$NE #"?;E9
9AF ,1F1G,/F&G,&F,
S.T.
17 ,F1G/F&G&F,J83
&7 1F1G2F&G8F,C1/3
,7 /F1G,F&G,F,C1&3
"#T$9A; S";<T$"N
1/ Chapter 8 LP Sensitivity Analysis and Interpretation of Solution
"b*ective @unction :alue B 21,.,,,
:ariable :alue educed %ost
F1 1,.,,, 3.333
F& 13.333 3.333
F, 3.333 13.888
%onstraint
Slac.ASurplu
s
+ual #rice
1 3.333 (3.228
& 3.333 /.//1
, &,.,,, 3.333
"?NE%T$:E %"E@@$%$ENT ANEES
:ariable ;ower ;imit %urrent :alue <pper ;imit
F1 ,3.333 ,1.333 No <pper ;imit
F& No ;ower ;imit ,/.333 ,1.112
F, No ;ower ;imit ,&.333 -&.888
$EDT DAN+ S$+E ANEES
%onstraint ;ower ;imit %urrent :alue <pper ;imit
1 22.1-2 83.333 132.1-,
& 1&1.333 1/3.333 11,.1&/
, 81.112 1&3.333 No <pper ;imit
a. Eive the solution to the problem.
b. Which constraints are binding=
c. What would happen if the coefficient of 51 increased b0 ,=
d. What would happen if the right(hand side of constraint 1 increased b0 13=
T"#$%! $nterpretation of 9anagement Scientist output
8. <se the following 9anagement Scientist output to answer the questions.
9$N -F1G/F&G1F,
S.T.
17 F1GF&GF,C8/
&7 ,F1G-F&G&F,J&83
,7 &F1G-F&G-F,J,&3
"b*ective @unction :alue B -33.333
:ariable :alue educed %ost
F1 3.333 1./33
F& 83.333 3.333
F, 3.333 1.333
%onstraint
Slac.ASurplu
s
+ual #rice
Chapter 8 LP Sensitivity Analysis and Interpretation of Solution 11
1 /.333 3.333
& -3.333 3.333
, 3.333 (1.&/3
"?NE%T$:E %"E@@$%$ENT ANEES
:ariable ;ower ;imit %urrent :alue <pper ;imit
F1 &./33 -.333 No <pper ;imit
F& 3.333 /.333 1.333
F, /.333 1.333 No <pper ;imit
$EDT DAN+ S$+E ANEES
%onstraint ;ower ;imit %urrent :alue <pper ;imit
1 83.333 8/.333 No <pper ;imit
& No ;ower ;imit &83.333 ,&3.333
, &83.333 ,&3.333 ,-3.333
a. What is the optimal solution) and what is the value of the profit contribution=
b. Which constraints are binding=
c. What are the dual prices for each resource= $nterpret.
d. %ompute and interpret the ranges of optimalit0.
e. %ompute and interpret the ranges of feasibilit0.
T"#$%! $nterpretation of 9anagement Scientist output
8. The following linear programming problem has been solved b0 The 9anagement Scientist. <se the
output to answer the questions.
;$NEA #"EA99$NE #"?;E9
9AF &/F1G,3F&G1/F,
S.T.
17 -F1G/F&G8F,C1&33
&7 8F1G1/F&G,F,C1/33
"#T$9A; S";<T$"N
"b*ective @unction :alue B -233.333
:ariable :alue educed %ost
F1 1-3.333 3.333
F& 3.333 13.333
F, 83.333 3.333
%onstraint
Slac.ASurplu
s
+ual #rice
1 3.333 1.333
& 3.333 &.,,,
12 Chapter 8 LP Sensitivity Analysis and Interpretation of Solution
"?NE%T$:E %"E@@$%$ENT ANEES
:ariable ;ower ;imit %urrent :alue <pper ;imit
F1 18.&81 &/.333 -/.333
F& No ;ower ;imit ,3.333 -3.333
F, 8.,,, 1/.333 /3.333
$EDT DAN+ S$+E ANEES
%onstraint ;ower ;imit %urrent :alue <pper ;imit
1 111.112 1&33.333 -333.333
& -/3.333 1/33.333 &233.333
a. Eive the complete optimal solution.
b. Which constraints are binding=
c. What is the dual price for the second constraint= What interpretation does this have=
d. "ver what range can the ob*ective function coefficient of 5& var0 before a new solution point
becomes optimal=
e. ?0 how much can the amount of resource & decrease before the dual price will change=
f. What would happen if the first constraintOs right(hand side increased b0 233 and the secondOs
decreased b0 ,/3=
T"#$%! $nterpretation of 9anagement Scientist output

13. ;$N+" output is given for the following linear programming problem.
9$N 1& F1 G 13 F& G 8 F,
S<?NE%T T"
&7 / F1 G 8 F& G / F, JB 13
,7 8 F1 G 13 F& G / F, JB 83
EN+

;# "#T$9<9 @"<N+ AT STE# 1
"?NE%T$:E @<N%T$"N :A;<E
17 83.333333
:A$A?;E :A;<E E+<%E+ %"ST
F1 .333333 -.333333
F& 8.333333 .333333
F, .333333 -.333333
"W S;A%K " S<#;<S +<A; #$%E
&7 -.333333 .333333
,7 .333333 (1.333333
N". $TEAT$"NSB 1
ANEES $N WD$%D TDE ?AS$S $S <N%DANEE+!

"?N. %"E@@$%$ENT ANEES
Chapter 8 LP Sensitivity Analysis and Interpretation of Solution 13
:A$A?;E
%<ENT
%"E@@$%$ENT
A;;"WA?;E
$N%EASE
A;;"WA?;E
+E%EASE
F1 1&.333333 $N@$N$TH -.333333
F& 13.333333 /.333333 13.333333
F, 8.333333 $N@$N$TH -.333333
$EDTDAN+ S$+E ANEES
"W
%<ENT
DS
A;;"WA?;E
$N%EASE
A;;"WA?;E
+E%EASE
& 13.333333 -.333333 $N@$N$TH
, 83.333333 $N@$N$TH /.333333
a. What is the solution to the problem=
b. Which constraints are binding=
c. $nterpret the reduced cost for 51.
d. $nterpret the dual price for constraint &.
e. What would happen if the cost of 51 dropped to 13 and the cost of 5& increased to 1&=
T"#$%! $nterpretation of ;$N+" output
11. The ;# problem whose output follows determines how man0 nec.laces) bracelets) rings) and earrings a
*ewelr0 store should stoc.. The ob*ective function measures profitP it is assumed that ever0 piece stoc.ed
will be sold. %onstraint 1 measures displa0 space in units) constraint & measures time to set up the displa0
in minutes. %onstraints , and - are mar.eting restrictions.
;$NEA #"EA99$NE #"?;E9
9AF 133F1G1&3F&G1/3F,G1&/F-
S.T.
17 F1G&F&G&F,G&F-C138
&7 ,F1G/F&GF-C1&3
,7 F1GF,C&/
-7 F&GF,GF-J/3
"#T$9A; S";<T$"N
"b*ective @unction :alue B 2-2/.333
:ariable :alue educed %ost
F1 8.333 3.333
F& 3.333 /.333
F, 12.333 3.333
F- ,,.333 3.333
%onstraint
Slac.ASurplu
s
+ual #rice
1 3.333 2/.333
& 1,.333 3.333
, 3.333 &/.333
- 3.333 (&/.333
"?NE%T$:E %"E@@$%$ENT ANEES
14 Chapter 8 LP Sensitivity Analysis and Interpretation of Solution
:ariable ;ower ;imit %urrent :alue <pper ;imit
F1 82./33 133.333 No <pper ;imit
F& No ;ower ;imit 1&3.333 1&/.333
F, 1&/.333 1/3.333 11&./33
F- 1&3.333 1&/.333 1/3.333
$EDT DAN+ S$+E ANEES
%onstraint ;ower ;imit %urrent :alue <pper ;imit
1 133.333 138.333 1&,.2/3
& /2.333 1&3.333 No <pper ;imit
, 8.333 &/.333 /8.333
- -1./33 /3.333 /-.333
<se the output to answer the questions.
a. Dow man0 nec.laces should be stoc.ed=
b. Now man0 bracelets should be stoc.ed=
c. Dow man0 rings should be stoc.ed=
d. Dow man0 earrings should be stoc.ed=
e. Dow much space will be left unused=
f. Dow much time will be used=
g. ?0 how much will the second mar.eting restriction be e5ceeded=
h. What is the profit=
i. To what value can the profit on nec.laces drop before the solution would change=
*. ?0 how much can the profit on rings increase before the solution would change=
.. ?0 how much can the amount of space decrease before there is a change in the profit=
l. Hou are offered the chance to obtain more space. The offer is for 1/ units and the total price is
1/33. What should 0ou do=
T"#$%! $nterpretation of 9anagement Scientist output
1&. The decision variables represent the amounts of ingredients 1) &) and , to put into a blend. The
ob*ective function represents profit. The first three constraints measure the usage and availabilit0
of resources A) ?) and %. The fourth constraint is a minimum requirement for ingredient ,. <se
the output to answer these questions.
a. Dow much of ingredient 1 will be put into the blend=
b. Dow much of ingredient & will be put into the blend=
c. Dow much of ingredient , will be put into the blend=
d. Dow much resource A is used=
e. Dow much resource ? will be left unused=
f. What will the profit be=
g. What will happen to the solution if the profit from ingredient & drops to -=
h. What will happen to the solution if the profit from ingredient , increases b0 1=
i. What will happen to the solution if the amount of resource % increases b0 &=
*. What will happen to the solution if the minimum requirement for ingredient , increases to
1/=
;$NEA #"EA99$NE #"?;E9
9AF -F1G1F&G2F,
Chapter 8 LP Sensitivity Analysis and Interpretation of Solution 15
S.T.
17 ,F1G&F&G/F,C1&3
&7 1F1G,F&G,F,C83
,7 /F1G/F&G8F,C113
-7 G1F,J13
"#T$9A; S";<T$"N
"b*ective @unction :alue B 111.333
:ariable :alue educed %ost
F1 3.333 &.333
F& 11.333 3.333
F, 13.333 3.333
%onstraint
Slac.ASurplu
s
+ual #rice
1 ,8.333 3.333
& &.333 3.333
, 3.333 1.&33
- 3.333 (&.133
"?NE%T$:E %"E@@$%$ENT ANEES
:ariable ;ower ;imit %urrent :alue <pper ;imit
F1 No ;ower ;imit -.333 1.333
F& -.,2/ 1.333 No <pper ;imit
F, No ;ower ;imit 2.333 8.133
$EDT DAN+ S$+E ANEES
%onstraint ;ower ;imit %urrent :alue <pper ;imit
1 8&.333 1&3.333 No <pper ;imit
& 28.333 83.333 No <pper ;imit
, 83.333 113.333 11,.,,,
- 8.888 13.333 &3.333
T"#$%! $nterpretation of 9anagement Scientist output
1,. The ;# model and ;$N+" output represent a problem whose solution will tell a specialt0 retailer how
man0 of four different st0les of umbrellas to stoc. in order to ma5imi'e profit. $t is assumed that ever0
one stoc.ed will be sold. The variables measure the number of womenOs) golf) menOs) and folding
umbrellas) respectivel0. The constraints measure storage space in units) special displa0 rac.s) demand)
and a mar.eting restriction) respectivel0.

9AF - F1 G 1 F& G / F, G ,./ F-
S<?NE%T T"
&7 & F1 G , F& G , F, G F- CB 1&3
,7 1./ F1 G & F& CB /-
-7 & F& G F, G F- CB 2&
/7 F& G F, JB 1&
EN+
16 Chapter 8 LP Sensitivity Analysis and Interpretation of Solution
"?NE%T$:E @<N%T$"N :A;<E
17 ,18.33333
:A$A?;E :A;<E E+<%E+ %"ST
F1 1&.333333 .333333
F& .333333 ./33333
F, 1&.333333 .333333
F- 13.333333 .333333
"W S;A%K " S<#;<S +<A; #$%E
&7 .333333 &.333333
,7 ,1.333333 .333333
-7 .333333 1./33333
/7 .333333 (&./33333
ANEES $N WD$%D TDE ?AS$S $S <N%DANEE+!

"?N. %"E@@$%$ENT ANEES
:A$A?;E
%<ENT
%"E@@$%$ENT
A;;"WA?;E
$N%EASE
A;;"WA?;E
+E%EASE
F1 -.333333 1.333333 &./33333
F& 1.333333 ./33333 $N@$N$TH
F, /.333333 &./33333 ./33333
F- ,./33333 $N@$N$TH ./33333
$EDTDAN+ S$+E ANEES
"W
%<ENT
DS
A;;"WA?;E
$N%EASE
A;;"WA?;E
+E%EASE
& 1&3.333333 -8.333333 &-.333333
, /-.333333 $N@$N$TH ,1.333333
- 2&.333333 &-.333333 -8.333333
/ 1&.333333 1&.333333 1&.333333
<se the output to answer the questions.
a. Dow man0 womenOs umbrellas should be stoc.ed=
b. Dow man0 golf umbrellas should be stoc.ed=
c. Dow man0 menOs umbrellas should be stoc.ed=
d. Dow man0 folding umbrellas should be stoc.ed=
e. Dow much space is left unused=
f. Dow man0 rac.s are used=
g. ?0 how much is the mar.eting restriction e5ceeded=
h. What is the total profit=
i. ?0 how much can the profit on womenOs umbrellas increase before the solution would change=
*. To what value can the profit on golf umbrellas increase before the solution would change=
.. ?0 how much can the amount of space increase before there is a change in the dual price=
l. Hou are offered an advertisement that should increase the demand constraint from 2& to 81 for a
total cost of M&3. Would 0ou sa0 0es or no=
T"#$%! $nterpretation of ;$N+" output
Chapter 8 LP Sensitivity Analysis and Interpretation of Solution 17
1-. Eight of the entries have been deleted from the ;$N+" output that follows. <se what 0ou .now about
linear programming to find values for the blan.s.
9$N 1 F1 G 2./ F& G 13 F,
S<?NE%T T"
&7 &/ F1 G ,/ F& G ,3 F, JB &-33
,7 & F1 G - F& G 8 F, JB -33
EN+

;# "#T$9<9 @"<N+ AT STE# &
"?NE%T$:E @<N%T$"N :A;<E
17 11&./3333
:A$A?;E :A;<E E+<%E+ %"ST
F1 IIIIIIII 1.,1&/33
F& IIIIIIII IIIIIIII
F, &2./33333 IIIIIIII
"W S;A%K " S<#;<S +<A; #$%E
&7 IIIIIIII (.1&/333
,7 IIIIIIII (.281&/3
N". $TEAT$"NSB &
ANEES $N WD$%D TDE ?AS$S $S <N%DANEE+!

"?N. %"E@@$%$ENT ANEES
:A$A?;E
%<ENT
%"E@@$%$ENT
A;;"WA?;E
$N%EASE
A;;"WA?;E
+E%EASE
F1 1.333333 IIIIIIIII IIIIIIIII
F& 2./33333 1./33333 &./33333
F, 13.333333 /.333333 ,./21-&8
$EDTDAN+ S$+E ANEES
"W
%<ENT
DS
A;;"WA?;E
$N%EASE
A;;"WA?;E
+E%EASE
& &-33.333333 1133.333333 833.333333
, -33.333333 &-3.333333 1&/.21-,33
T"#$%! $nterpretation of ;$N+" output
1/. #ortions of a 9anagement Scientist output are shown below. <se what 0ou .now about the solution of
linear programs to fill in the ten blan.s.
;$NEA #"EA99$NE #"?;E9
9AF 1&F1G8F&G2F,
S.T.
17 ,F1G/F&G-F,C1/3
1% Chapter 8 LP Sensitivity Analysis and Interpretation of Solution
&7 &F1G1F&G1F,C1-
,7 1F1G&F&G1F,C83
-7 &F1G-F&G,F,J111
"#T$9A; S";<T$"N
"b*ective @unction :alue B ,,1.333
:ariable :alue educed %ost
F1 IIIIII 3.333
F& &-.333 IIIIII
F, IIIIII ,./33
%onstraint
Slac.ASurplu
s
+ual #rice
1 3.333 1/.333
& IIIIII 3.333
, IIIIII 3.333
- 3.333 IIIIII
"?NE%T$:E %"E@@$%$ENT ANEES
:ariable ;ower ;imit %urrent :alue <pper ;imit
F1 /.-33 1&.333 No <pper ;imit
F& &.333 8.333 &3.333
F, No ;ower ;imit 2.333 13./33
$EDT DAN+ S$+E ANEES
%onstraint ;ower ;imit %urrent :alue <pper ;imit
1 1-/.333 1/3.333 1/1.112
& IIIIII IIIIII 1-.333
, IIIIII IIIIII 83.333
- 113.&81 111.333 1&3.333
T"#$%! $nterpretation of 9anagement Scientist output
Note to $nstructor3 The following problem is suitable for a ta.e(home or lab e5am. The student must formulate
the model) solve the problem with a computer pac.age) and then interpret the solution to answer the questions.
11. A large sporting goods store is placing an order for bic0cles with its supplier. @our models can be
ordered! the adult "pen Trail) the adult %it0scape) the girlOs Sea Sprite) and the bo0Os Trail ?la'er. $t is
assumed that ever0 bi.e ordered will be sold) and their profits) respectivel0) are ,3) &/) &&) and &3. The
;# model should ma5imi'e profit. There are several conditions that the store needs to worr0 about. "ne
of these is space to hold the inventor0. An adult>s bi.e needs two feet) but a childOs bi.e needs onl0 one
foot. The store has /33 feet of space. There are 1&33 hours of assembl0 time available. The childOs bi.e
need - hours of assembl0 timeP the "pen Trail needs / hours and the %it0scape needs 1 hours. The store
would li.e to place an order for at least &2/ bi.es.
a. @ormulate a model for this problem.
b. Solve 0our model with an0 computer pac.age available to 0ou.
c. Dow man0 of each .ind of bi.e should be ordered and what will the profit be=
d. What would the profit be if the store had 133 more feet of storage space=
Chapter 8 LP Sensitivity Analysis and Interpretation of Solution 1*
e. $f the profit on the %it0scape increases to M,/) will an0 of the %it0scape bi.es be ordered=
f. "ver what range of assembl0 hours is the dual price applicable=
g. $f we require / more bi.es in inventor0) what will happen to the value of the optimal solution=
h. Which resource should the compan0 wor. to increase) inventor0 space or assembl0 time=
T"#$%! @ormulation and computer solution
12. A compan0 produces two products made from aluminum and copper. The table below gives the unit
requirements) the unit production man(hours required) the unit profit and the availabilit0 of the resources
6in tons7.
Aluminum %opper 9an(hours <nit #rofit
#roduct 1 1 3 & /3
#roduct & 1 1 , 13
Available 13 1 &-
The Management Scientist provided the following solution output!
"?NE%T$:E @<N%T$"N :A;<E B /-3.333
:A$A?;E :A;<E E+<%E+ %"ST
F1 1.333 3.333
F& -.333 3.333
%"NSTA$NT S;A%KAS<#;<S +<A; #$%E
1 .333 ,3.333
& &.333 3.333
, 3.333 13.333
ANEES $N WD$%D TDE ?AS$S $S <N%DANEE+!

"?N. %"E@@$%$ENT ANEES
:A$A?;E
%<ENT
%"E@@$%$ENT
A;;"WA?;E
$N%EASE
A;;"WA?;E
+E%EASE
F1 /3.333 13.333 13.333
F& 13.333 1/.333 13.333
$EDTDAN+ S$+E ANEES
%"NSTA$NT
%<ENT
DS
A;;"WA?;E
$N%EASE
A;;"WA?;E
+E%EASE
1 13.333 &.333 1.333
& 1.333 $N@$N$TH &.333
, &-.333 &.333 -.333
a. What is the optimal production schedule=
b. Within what range for the profit on product & will the solution in 6a7 remain optimal= What is
the optimal profit when c
&
B 23=
c. Suppose that simultaneousl0 the unit profits on x
1
and x
&
changed from /3 to // and 13 to 1/
respectivel0. Would the optimal solution change=
d. E5plain the meaning of the Q+<A; #$%ESQ column. Eiven the optimal solution) wh0 should
the dual price for copper be 3=
e. What is the increase in the value of the ob*ective function for an e5tra unit of aluminum=
2/ Chapter 8 LP Sensitivity Analysis and Interpretation of Solution
f. 9an(hours were not figured into the unit profit as it must pa0 three wor.ers for eight hours of
wor. regardless of the number of man(hours used. What is the dual price for man(hours=
$nterpret.
g. "n the other hand) aluminum and copper are resources that are ordered as needed. The unit
profit coefficients were determined b0! 6selling price per unit7 ( 6cost of the resources per unit7.
The 13 units of aluminum cost the compan0 M133. What is the most the compan0 should be
willing to pa0 for e5tra aluminum=
T"#$%! $nterpretation of solution
18. Eiven the following linear program!
9AF /x
1
G 2x
&
s.t. x
1
C 1
&x
1
G ,x
&
C 18
x
1
G x
&
C 8
x
1
) x
&
J 3
The graphical solution to the problem is shown below. @rom the graph we see that the optimal solution
occurs at x
1
B /) x
&
B ,) and z B -1.

8
7
6
5
4
3
2
1
1 2 3 4 5 6 7 8 9 10
2X1 + 3X2 < 19
X2
X1
X1 + X2 < 8
MAX 5X1 + 7X2
X1 < 6
Optimal X1 = 5, X2 = 3
Z = 46
a. %alculate the range of optimalit0 for each ob*ective function coefficient.
b. %alculate the dual price for each resource.
T"#$%! $ntroduction to sensitivit0 anal0sis
18. %onsider the following linear program!
9AF ,x
1
G -x
&
6M #rofit7
Chapter 8 LP Sensitivity Analysis and Interpretation of Solution 21
s.t. x
1
G ,x
&
C 1&
&x
1
G x
&
C 8
x
1
C ,
x
1
) x
&
J 3
The Management Scientist provided the following solution output!

"#T$9A; S";<T$"N
"b*ective @unction :alue B &3.333
:ariable :alue educed %ost
F1 &.-33 3.333
F& ,.&33 3.333
%onstraint
Slac.ASurplu
s
+ual #rice
1 3.333 1.333
& 3.333 1.333
, 3.133 3.333
"?NE%T$:E %"E@@$%$ENT ANEES
:ariable ;ower ;imit %urrent :alue <pper ;imit
F1 1.,,, ,.333 8.333
F& 1./33 -.333 8.333
$EDT DAN+ S$+E ANEES
%onstraint ;ower ;imit %urrent :alue <pper ;imit
1 8.333 1&.333 &-.333
& -.333 8.333 8.333
, &.-33 ,.333 No <pper ;imit
a. What is the optimal solution including the optimal value of the ob*ective function=
b. Suppose the profit on x
1
is increased to M2. $s the above solution still optimal= What is the value
of the ob*ective function when this unit profit is increased to M2=
c. $f the unit profit on x
&
was M13 instead of M-) would the optimal solution change=
d. $f simultaneousl0 the profit on x
1
was raised to M/./ and the profit on x
&
was reduced to M,)
would the current solution still remain optimal=
T"#$%! $nterpretation of solution
&3. %onsider the following linear program!
9$N 1x
1
G 8x
&
6M cost7
s.t. x
1
G &x
&
C 8
13x
1
G 2./x
&
J ,3
x
&
J &
x
1
) x
&
J 3
22 Chapter 8 LP Sensitivity Analysis and Interpretation of Solution
The Management Scientist provided the following solution output!
"#T$9A; S";<T$"N
"b*ective @unction :alue B &2.333
:ariable :alue educed %ost
F1 1./33 3.333
F& &.333 3.333
%onstraint
Slac.ASurplu
s
+ual #rice
1 &./33 3.333
& 3.333 (3.133
, 3.333 (-./33
"?NE%T$:E %"E@@$%$ENT ANEES
:ariable ;ower ;imit %urrent :alue <pper ;imit
F1 3.333 1.333 1&.333
F& -./33 8.333 No <pper ;imit
$EDT DAN+ S$+E ANEES
%onstraint ;ower ;imit %urrent :alue <pper ;imit
1 /./33 8.333 No <pper ;imit
& 1/.333 ,3.333 //.333
, 3.333 &.333 -.333
a. What is the optimal solution including the optimal value of the ob*ective function=
b. Suppose the unit cost of x
1
is decreased to M-. $s the above solution still optimal= What is the
value of the ob*ective function when this unit cost is decreased to M-=
c. Dow much can the unit cost of x
&
be decreased without concern for the optimal solution
changing=
d. $f simultaneousl0 the cost of 5
1
was raised to M2./ and the cost of x
&
was reduced to M1) would
the current solution still remain optimal=
e. $f the right(hand side of constraint , is increased b0 1) what will be the effect on the optimal
solution=
T"#$%! $nterpretation of solution
SOLUTIO"S TO PRO$LEMS
1. a. (/A, and (&A/
b. "b*ective functions &7) ,7) and -7
&. a. 1.,, < c1 < -
b. . / < c& < 1./
c. +ual prices are .&/) .&/) 3
Chapter 8 LP Sensitivity Analysis and Interpretation of Solution 23
,. a. .8 < c1 < &
b. 1 < c& < &./
c. 51 B &/3) 5& B /3) ' B -2/
d. %onstraints 1 and & will be binding.
e. +ual prices are .,,) 3) .,, 6The first and third values are negative.7
-. a. 9a5 -F G 1H
s.t. ,F G /H C 13
,F G &H C -8
1F G 1H C &3
F ) H J 3
b. The complete optimal solution is F B 1,.,,,) H B -) R B 2,.,,,) S1 B 3) S& B 3) S, B &.112
/. a. 9in /F G -H
s.t. -F G ,H J 13
&F G /H J /3
8F G 8H J 1--
F ) H J 3
b. The complete optimal solution is F B 8.1) H B 2.&) R B 21.8) S1 B 3) S& B /.&) S, B 3
1. a. B?8S?11
b. B%8S%11
c. B?1&G%1&
d. B?-S?11G%-S%11
e. B?/S?11G%/S%11
f. B?1S?11G%1S%11
g. 8.-1
h. -.11
i. 0es
*. no
2. a. 51 B 1,.,,) 5& B 13) 5, B 3) s1 B 3) s& B 3) s, B &,.,,) ' B 21,.,,
b. %onstraints 1 and & are binding.
c. The value of the ob*ective function would increase b0 -3.
d. The value of the ob*ective function would decrease b0 2.28.
8. a. 51 B 3) 5& B 83) 5, B 3) s1 B /) s& B -3) s, B 3) R B -33
b. %onstraint , is binding.
c. +ual prices are 3) 3) and (1.&/.
The0 measure the improvement in R per unit increase in each right(hand side.
d. &./ C c1 C
3 C c& C 1
/ C c, C
As long as the ob*ective function coefficient sta0s within its range) the current optimal solution
point will not change) although R could.
e. 83 C b1 C
( C b& C ,&3
&83 C b, C ,-3
As long as the right(hand side value sta0s within its range) the currentl0 binding constraints will
remain so) although the values of the decision variables could change. The dual variable values
will remain the same.
8. a. 51 B 1-3) 5& B 3) 5, B 83) s1 B 3) s& B 3) ' B -233
b. %onstraints 1 and & are binding.
24 Chapter 8 LP Sensitivity Analysis and Interpretation of Solution
c. +ual price & B &.,,. A unit increase in the right(hand side of constraint & will increase the value
of the ob*ective function b0 &.,,.
d. As long as c& C -3) the solution will be unchanged.
e. 13/3
f. The sum of percentage changes is 233A&833 G 6(,/37A6(13/37 C 1 so the solution will not change.
13. a. 51 B 3) 5& B 8) 5, B 3) s1 B -) s& B 3) ' B 83
b. %onstraint & is binding.
c. c1 would have to decrease b0 - or more for 51 to become positive.
d. $ncreasing the right(hand side b0 1 will cause a negative improvement) or increase) of 1 in this
minimi'ation ob*ective function.
e. The sum of the percentage changes is 6(&7A6(-7 G &A/ C 1 so the solution would not change.
11. a. 8
b. 3
c. 12
d. ,,
e. 3
f. /2
g. 3
h. 2-2/
i. 82./
* 1&./
.. 3
l. Sa0 no. Although 1/ units can be evaluated) their value 611&/7 is less than the cost 61/337.
1&. a. 3
b. 11
c. 13
d. --
e. &
f. 111
g. rerun
h. R B 121
i. R B 118.-
*. R B 1/,
1,. a. 1&
b. 3
c. 1&
d. 13
e. 3
f. 18
g. 3
h. ,18
i. 1
*. 1./
.. -8
l. Hes. The dual price is 1./ for &- additional units. The value of the ad 61-761./7B&1 e5ceeds the
cost of &3.
1-. $t is easiest to calculate the values in this order.
51 B 3) 5& B -/) reduced cost & B 3) reduced cost , B 3) row & slac. B 3) row , slac. B 3) c1 allowable
decrease B 1.,1&/) allowable increase B infinit0
1/. 5, B 3 because the reduced cost is positive.
51 B &- after plugging into the ob*ective function
The second reduced cost is 3.
Chapter 8 LP Sensitivity Analysis and Interpretation of Solution 25
s& B &3 and s, B && from plugging into the constraints.
The fourth dual price is (11./ from plugging into the dual ob*ective function) which 0our students might
not understand full0 until %hapter 1.
The lower limit for constraint & is -- and for constraint , is /8) from the amount of slac. in each
constraint. There are no upper limits for these constraints.
11. a.
9AF ,3 F1 G &/ F& G && F, G &3 F-
S<?NE%T T"
&7 & F1 G & F& G F, G F- CB /33
,7 / F1 G 1 F& G - F, G - F- CB 1&33
-7 F1 G F& G F, G F- JB &2/

b.
"?NE%T$:E @<N%T$"N :A;<E
17 18/3.3333
:A$A?;E :A;<E E+<%E+ %"ST
F1 133.333333 .333333
F& .333333 1,.333333
F, 12/.333333 .333333
F- .333333 &.333333
"W S;A%K " S<#;<S +<A; #$%E
&7 1&/.333333 .333333
,7 .333333 8.333333
-7 .333333 (13.333333
N". $TEAT$"NSB &
ANEES $N WD$%D TDE ?AS$S $S <N%DANEE+!

"?N. %"E@@$%$ENT ANEES
:A$A?;E
%<ENT
%"E@@$%$ENT
A;;"WA?;E
$N%EASE
A;;"WA?;E
+E%EASE
F1 ,3.333333 $N@$N$TH &./33333
F& &/.333333 1,.333333 $N@$N$TH
F, &&.333333 &.333333 &.333333
F- &3.333333 &.333333 $N@$N$TH
$EDTDAN+ S$+E ANEES
"W
%<ENT
DS
A;;"WA?;E
$N%EASE
A;;"WA?;E
+E%EASE
& /33.333333 $N@$N$TH 1&/.333333
, 1&33.333333 1&/.333333 133.333333
- &2/.333333 &/.333333 ,/.333333
c. "rder 133 "pen Trails) 3 %it0scapes) 12/ Sea Sprites) and 3 Trail ?la'ers. #rofit will be 18/3.
d. 18/3
e. No. The M13 increase is below the reduced cost.
f. 1133 to 1,&/
26 Chapter 8 LP Sensitivity Analysis and Interpretation of Solution
g. $t will decrease b0 /3.
h. Assembl0 time.
12. a. 1 product 1) - product &) #rofit B M/-3
b. ?etween M/3 and M2/P at M23 the profit is M/83
c. NoP total 4 change is 8, 1A,4 C 1334
d. +ual prices are the shadow prices for the resourcesP since there was unused copper 6because S
&
B
&7) e5tra copper is worth M3
e. M,3
f. M13P this is the amount e5tra man(hours are worth
g. The shadow price is the QpremiumQ for aluminum (( would be willing to pa0 up to M13 G M,3 B
M-3 for e5tra aluminum
18. a. anges of optimalit0! 1-A, C c
1
C 2 and / C c
&
C 1/A&
b. Summari'ing) the dual price for the first resource is 3) for the second resource is &) and for the
third is 1
18. a. x
1
B &.- and x
&
B ,.&) and z B M&3.33.
b. "ptimal solution will not change. "ptimal profit will equal M&8.13.
c. ?ecause 13 is outside the range of 1./ to 8.3) the optimal solution li.el0 would change.
d. Sum of the change percentages is /34 G -34 B 834. Since this does not e5ceed 1334 the
optimal solution would not change.
&3. a. x
1
B 1./ and x
&
B &.3) and the ob*ective function value B &2.33.
b. - is within this range of 3 to 1&) so the optimal solution will not change. "ptimal total cost will
be M&-.33.
c. x
&
can fall to -./ without concern for the optimal solution changing.
d. Sum of the change percentages is 81.24. This does not e5ceed 1334) so the optimal solution
would not change.
e. The right(hand side remains within the range of feasibilit0) so there is no change in the optimal
solution. Dowever) the ob*ective function value increases b0 M-./3.