Professional Documents
Culture Documents
In this section we discuss applications in After considering possible advertising media and the
media selection and marketing research. market to be covered, BP&J recom-
mended that the first month’s advertising be restricted
to five media. At the end of the month,
Media Selection
BP&J will then reevaluate its strategy based on the
Media selection applications of linear month’s results. BP&J collected data
programming are designed to help on the number of potential customers reached, the
cost per advertisement, the maximum
marketing managers allocate a fixed
number of times each medium is available, and the
advertising budget to various advertising exposure quality rating for each of the
media. five media. The quality rating is measured in terms of
an exposure quality unit, a measure of
the relative value of one advertisement in each of the
media. This measure, based on BP&J’s
experience in the advertising business, takes into The data on quality of exposure in Table
account factors such as audience demographics (age,
Show that each daytime TV (DTV)
income, and education of the audience reached),
image presented, and quality of the advertisement.
advertisement is rated at 65 exposure
quality units.
Relax-and-Enjoy provided BP&J with an advertising
budget of $30,000 for the first an advertising plan with DTV
month’s campaign.
advertisements will provide a total of 65DTV
In addition, Relax-and-Enjoy imposed the following exposure quality units.
restrictions on how BP&J may allocate these funds:
Continuing with the data in Table 4.1, we
At least 10 television commercials must be used, at
find evening TV (ETV) rated at 90 exposure
least 50,000 potential customers must be reached,
and no more than $18,000 may be spent on
quality units, daily newspaper (DN) rated at
television advertisements. 40 exposure quality units
NOTES:
What portfolio
recommendations—investments and
amounts—should be made for
the available $100,000? Given the objective
of maximizing projected return subject to the
Managers of mutual funds, credit unions, budgetary and managerial imposed
insurance companies, and banks frequently constraints, we can answer this question by
encounter this type of problem. The formulating and solving a linear
objective function for portfolio selection programming model of the problem. The
problems usually is maximization of solution will provide investment
expected return or minimization of risk. The recommendations for the management of
constraints usually take the form of Welte Mutual Funds.
restrictions on the type of permissible
investments, state laws, company policy,
maximum permissible risk, and so on. A =dollars invested in Atlantic Oil
Problems of this type have been formulated P =dollars invested in Pacific Oil
and solved using a variety of mathematical M =dollars invested in Midwest Steel
programming techniques. In this section we H =dollars invested in Huber Steel
formulate and G =dollars invested in government bonds
solve a portfolio selection problem as a
linear program. we write the objective function for
maximizing the total return for the portfolio
as
Management of Welte imposed the Max 0.073A + 0.103P + 0.064M +
following investment guidelines: 0.075H + 0.045G
1. Neither industry (oil or steel) should
receive more than $50,000. The constraint specifying investment of the
2. Government bonds should be at least available $100,000 is
25% of the steel industry investments. A + P + M + H + G = 100,000
3. The investment in Pacific Oil, the
high-return but high-risk investment, cannot The requirements that neither the oil nor the
be steel industry should receive more than
more than 60% of the total oil industry $50,000 are
investment. A + P ≤ 50 000
M + H ≤ 50 000
The requirement that government bonds be The optimal solution to this linear program is
at least 25% of the steel industry investment shown in Figure 4.3. Table 4.4 shows how
is expressed as The funds are divided among the securities.
G ≥ 0.25 (M + H) Note that the optimal solution indicates that
the portfolio should be diversified among all
Finally, the constraint that Pacific Oil cannot the investment opportunities except
be more than 60% of the total oil industry Midwest Steel. The projected annual return
investment is for this portfolio is $8000, which is an overall
P ≤ 0.60 (A + P) return of 8%.
The optimal solution shows the dual value
for constraint 3 is zero. The reason is that
By adding the nonnegativity restrictions, we the steel industry maximum isn’t a binding
obtain the complete linear programming constraint; increases in the steel industry
model for the Welte Mutual Funds limit of $50,000 will not improve the value of
investment problem: the optimal solution. Indeed, the slack
variable for this constraint shows that the
Max current steel industry investment is $10,000
below its limit of $50,000. The dual values
0.073A + 0.103P + 0.064M + 0.075H + for the other constraints are nonzero,
0.045G indicating that these constraints are binding.
s.t. The dual value of 0.069 for constraint 1
shows that the value of the optimal solution
A + P + M + H + G = 100,000 can be increased by 0.069 if one more
Available funds dollar can be made available for the
A + P ≤ 50,000 portfolio investment. If more funds can be
Oil industry obtained at a cost of less than 6.9%,
maximum management should consider obtaining
them.
M + H ≤ 50,000
Steel industry
maximum
G ≥ 0.25 (M + H)
Government
bonds minimum
P ≤ 0.60 (A + P)
Pacific Oil Restriction
A, P, M, H, G ≥ 0
However, if a return in excess of 6.9% can With such an increase the Midwest Steel
be obtained by investing funds return would be 0.064 + 0.011 = 0.075,
elsewhere (other than in these five making this investment just as desir-
securities), management should question able as the currently used Huber Steel
the wisdom of investing the entire $100,000 investment alternative.
in this portfolio. Similar interpretations can
be given to the other dual values. Note that
the dual value for constraint 4 is negative at
–0.024. This result indicates that increasing
the value on the right-hand side of the Finally, a simple modification of the Welte
constraint by one unit can be expected to linear programming model permits
decrease the objective function value of the determining the fraction of available funds
optimal solution by 0.024. In terms of the invested in each security. That is, we divide
optimal portfolio, then, if Welte invests one each of the right-hand-side values by
more dollar in government bonds (beyond 100,000. Then the optimal values for the
the minimum requirement), the total variables will give the fraction of funds that
return will decrease by $0.024. To see why should be invested in each security for a
this decrease occurs, note again from the portfolio of any size.
dual value for constraint 1 that the marginal
return on the funds invested in the portfolio
is 6.9% (the average return is 8%). The rate
of return on government bonds is 4.5%.
Financial Planning
Thus, the cost of investing one more dollar Linear programming has been used for a
in government bonds is the difference variety of financial planning applications.
between the marginal The Management Science in Action,
return on the portfolio and the marginal General Electric Uses Linear Programming
return on government bonds: 6.9% – 4.5% 5 for Solar Energy Investment Decisions,
2.4%. Note that the optimal solution shows describes how linear programming is used
that Midwest Steel should not be included in to evaluate various scenarios to guide
the portfolio (M 5 0). The associated capital investment strategy over a long-term
reduced cost for M of –0.011 tells us that horizon.
the objective function coefficient for Midwest
Steel would have to increase by 0.011
before considering the Midwest Steel
investment alternative would be advisable.
In the rest of this section, we describe an A key feature of this type of financial
application of linear programming to planning problem is that a constraint must
minimize the cost of satisfying a company’s be formulated for each year of the planning
obligations to its early retirement program. horizon. In general, each constraint takes
the form:
Min F
Note that the constraint for year 6 shows The solution also shows that $636,148 (see
that funds available from bond 1 are S1) will be placed in savings at the
1.08875B1. The coefficient of 1.08875 beginning of the first year. By starting with
reflects the fact that bond 1 matures at the $1,728,794, the company can make the
end of year 5. As a result, the par value plus specified bond and savings investments and
the interest from bond 1 during year 5 is have enough leftover to meet the retirement
available at the beginning of year 6. Also, program’s first-year cash requirement of
because bond 1 matures in year 5 and $430,000. The optimal solution in Figure 4.4
becomes available for use at the beginning shows that the decision variables S1, S2,
of year 6, the variable B1 does not appear S3, and S4 all are greater than zero,
in the constraints for years 7 and 8. Note indicating investments in savings are
the similar interpretation for bond 2, which required in each of the first four
matures at the end of year 6 and has the years. However, interest from the bonds
par value plus interest available at the plus the bond maturity incomes will be
beginning of year 7. In addition, sufficient to cover the retirement program’s
bond 3 matures at the end of year 7 and cash requirements in years 5 through 8.
has the par value plus interest available at The dual values have an interesting
the beginning of year 8. Finally, note that a interpretation in this application. Each
variable S8 appears in the constraint for right-hand-side value corresponds to the
year 8. The retirement fund obligation will payment that must be made in that year.
be completed at the beginning of year 8, so Note that the dual values are
we anticipate that S8 will be zero and no positive, indicating that increasing the
funds will be put into savings. However, the required payment in any year by $1000
formulation includes S8 in the event that would increase the total funds required for
the bond income plus interest from the the retirement program’s obligation by
savings in year 7 exceed the 255 cash $1000 times the dual value. Also note that
requirement for year 8. Thus, S8 is a the dual values show that increases in
surplus variable that shows any funds required payments in the early years have
remaining after the eight-year cash the largest impact. This makes sense in that
requirements have been satisfied. The there is little time to build up investment
optimal solution to this 12-variable, income in the early years versus the
8-constraint linear program is shown in subsequent years. This suggests that if
Figure 4.4. With an objective function value Hewlitt faces increases in required
of 1728.79385, the total investment required payments it would benefit by deferring those
to meet the retirement plan’s eight-year increases to later years if possible.
obligation is $1,728,794. Using the current
prices of $1150, $1000, and $1350 for each
of the bonds, respectively, we can
summarize the initial investments in the
three bonds as follows:
Operations Management Applications Company forecasters indicate that 3000
Financial Manager calculators and 2000
Linear programming applications developed Technician calculators will be needed.
for production and operations management However, manufacturing capacity is limited.
include scheduling, staffing, inventory The company has 200 hours of regular
control, and capacity planning. manufacturing time and 50 hours of
overtime that can be scheduled for the
In this section we describe examples with calculators. Overtime involves a premium at
make-or-buy decisions, production the additional cost of $9 per hour.
scheduling, and workforce assignments.
Table 4.6 shows manufacturing times (in
minutes) for the components.
A Make-or-Buy Decision
BM + BP = 5000 Bases
FCM + FCP = 3000 Financial cartridges The optimal solution to this 11-variable,
TCM + TCP = 2000 Technician cartridges 7-constraint linear program is shown in
FTM + FTP = 3000 Financial tops Figure 4.5. The optimal solution indicates
TTM + TTP = 2000 Technician tops that all 5000 bases (BM), 667 Financial
Manager cartridges (FCM), and 2000
Two constraints are needed to guarantee Technician cartridges (TCM) should be
that manufacturing capacities for regular manufactured. The remaining 2333
time and overtime cannot be exceeded. The Financial Manager cartridges (FCP), all the
first constraint limits overtime capacity to Financial Manager tops (FTP), and all
50 hours, or Technician tops (TTP) should be purchased.
No overtime manufacturing is necessary,
OT ≤ 50 and the total cost associated with the
optimal make-or-buy plan is $24,443.33.
Sensitivity analysis provides some
additional information about the unused
overtime capacity. The Reduced Costs
column shows that the overtime (OT)
premium would have to decrease by $4 per
hour before overtime production should be
considered. That is, if the overtime premium
is $9 2 $4 5 $5 or less, Janders may want to
replace some of the purchased components
with components manufactured on
overtime.
After the order is processed, a demand Because the production cost per unit is the
statement is sent to the production control same each month, we don’t need to include
department. The production control the production costs in the objective
department must then develop a function; that is, regardless of the
three-month production plan for the production schedule selected, the total
components. production cost will remain the same. In
other words, production costs are not
In arriving at the desired schedule, the relevant costs for the production scheduling
production manager will want to decision under consideration. In cases in
identify the following: which the production cost per unit is
expected to change each month, the
1. Total production cost variable production costs per unit per month
2. Inventory holding cost must be included in the objective function.
3. Change-in-production-level costs The solution for the Bollinger Electronics
problem will be the same regardless of
whether these costs are included;therefore,
we included them so that the value of the After estimating the effects of employee
linear programming objective function layoffs, turnovers, reassignment training
will include all the costs associated with the costs, and other costs associated with
problem. fluctuating production levels, Bollinger
estimates that the cost associated with
To incorporate the relevant inventory increasing the production level for any
holding costs into the model, we let sim month is $0.50 per unit increase. A similar
denote the inventory level for product i at cost associated with decreasing the
the end of month m. Bollinger determined production level for any month is $0.20 per
that on a monthly basis inventory holding unit. Thus, we write the third portion of the
costs are 1.5% of the cost of the product; objective function as:
that is, (0.015)($20) = $0.30 per unit for
component 322A and (0.015)($10) 5 $0.15 Change-in-production-level costs = 0.50I1 +
per unit for component 802B. A common 0.50I2 + 0.50I3 + 0.20D1 + 0.20D2 +
assumption made in using the linear 0.20D3
programming approach to production
scheduling is that monthly ending
inventories are an acceptable approximation Note that the cost associated with changes
to the average inventory levels throughout in production level is a function of the
the month. Making this assumption, we change in the total number of units
write the inventory holding cost portion of produced in month m compared to the total
the objective function as number of units produced in month m – 1. In
other production scheduling applications,
Inventory holding cost = 0.30s11 + 0.30s12 fluctuations in production level might be
+ 0.30s13 + 0.15s21 + 0.15s22 + 0.15s23 measured in terms of machine hours or
labor-hours required rather
To incorporate the costs of fluctuations in than in terms of the total number of units
production levels from month to month, we produced.
need to define two additional variables:
Combining all three costs, the complete
Im =increase in the total production level objective function becomes
necessary during month m
Dm = decrease in the total production level Min 20x11 + 20x12 + 20x13 + 10x21 +
necessary during month m 10x22 + 10x23 + 0.30s11+ 0.30s12 +
0.30s13 + 0.15s21 + 0.15s22 + 0.15s23 +
0.50I1+ 0.50I2 + 0.50I3 + 0.20D1 + 0.20D2
+ 0.20D3
We now consider the constraints. First, we
must guarantee that the schedule meets Month 3
customer demand. Because the units
shipped can come from the current month’s s12 + x13 - s13 = 5000
production or from inventory carried over s22 + x23 - s23 = 3000
from previous months, the demand
requirement takes the form
If the company specifies a minimum
(Ending inventory from previous month ) + inventory level at the end of the three-month
(Current production) - (Ending inventory for period of at least 400 units of component
this month ) = ( This month’s demand ) 322A and at least 200 units of component
802B, we can add the constraints
0.65P1 + 0.95P2 ≤ b1
0.45P1 + 0.85P2 ≤ b2
1.00P1 + 0.70P2 ≤ b3
Suppose that McCormick has a 0.15P1 + 0.30P2 ≤ b4
cross-training program that enables some
employees to be transferred between The labor-hours ultimately allocated to each
departments. By taking advantage of the department must be determined by a series
cross-training skills, a limited number of of labor balance equations, or constraints,
employees and labor-hours may be that include the number of hours initially
transferred from one department to another. assigned to each department plus the
number of hours transferred into the
For example, suppose that the department minus the number of hours
cross-training permits transfers as shown in transferred out of the department.
Table 4.12. Row 1 of this table shows that
some employees assigned to department 1 Using department 1 as an example, We
have cross-training skills that permit them to determine the workforce allocation as
be transferred to department 2 or 3. The follows:
right-hand column shows that, for the
current production planning period, a b1 = (Hours initially in department 1) +
maximum of 400 hours can be transferred (Hours transferred into department 1) -
from department 1. Similar cross-training (Hours transferred out of department 1)
transfer capabilities and capacities are
shown for departments 2, 3, and 4.
When workforce assignments are flexible,
we do not automatically know how many
hours of labor should be assigned to or be
transferred from each department.
Table 4.11 shows 6500 hours initially The additional constraints are:
assigned to department 1. We use the
transfer decision variables ti1 to denote t12 + t13 ≤ 400
transfers into department 1 and t1j to t23 + t24 ≤ 800
denote transfers from department 1. Table t34 ≤ 100
4.12 shows that the cross-training t41 + t42 ≤ 200
capabilities involving department 1 are
restricted to transfers from department 4 The complete linear programming model
(variable t41) and transfers to either has two product decision variables (P1 and
department 2 or department 3 (variables t12 P2),four department workforce assignment
and t13). variables (b1, b2, b3, and b4), seven
transfer variables (t12, t13, t23, t24, t34,
t41, and t42), and 12 constraints. Figure 4.8
Thus, we can express the total workforce shows the optimal solution to this linear
allocation for department 1 as: program. McCormick’s profit can be
increased by $84,011 2 $73,590 5 $10,421
b1 = 6500 + t41 - t12 - t13 by taking advantage of cross-training and
workforce transfers. The optimal product
Moving the decision variables for the mix of 6825 units of product 1 and 1751
workforce transfers to the left-hand side, units of product 2 can be achieved if t13 5
400 hours are transferred from department
we have the labor balance equation or 1 to department 3; t23 5 651 hours are
constraint transferred from department 2 to
department 3; and t24 5 149 hours are
b1 - t41 + t12 + t13 = 6500 transferred from department 2 to
department 4. The resulting workforce
This form of constraint will be needed for assignments for departments 1 through 4
each of the four departments. would provide 6100, 5200, 8051, and 1549
hours, respectively. If a manager has the
Thus, the following labor balance flexibility to assign personnel to different
constraints for departments 2, 3, and 4 departments, reduced workforce idle time,
would be added to the model: improved workforce utilization, and
improved profit should result. The linear
b2 - t12 - t42 + t23 + t24 = 6000 programming model in this section
b3 - t13 - t23 + t34 = 7000 automatically assigns employees and
b4 - t24 - t34 + t41 + t42 = 1400 labor-hours to the departments in the most
profitable manner.
Blending Problems The Grand Strand Oil Company produces
regular and premium gasoline for
Blending problems arise whenever a independent service stations in the
manager must decide how to blend two or southeastern United States. The Grand
more resources to produce one or more Strand refinery manufactures the gasoline
products. products by blending three petroleum
components. The gasoline is sold at
In these situations, the resources contain different prices, and the petroleum
one or more essential ingredients that must components have different costs. The firm
be blended into final products that will wants to determine how to mix or blend the
contain specific percentages of each. three components into the two gasoline
products and maximize profits.
In most of these applications, then, Data available show that regular gasoline
management must decide how much of can be sold for $2.90 per gallon and
each resource to purchase to satisfy premium gasoline for $3.00 per gallon. For
product specifications and product demands the current production planning period,
at minimum cost. Grand Strand can obtain the three
petroleum components at the cost per
Blending problems occur frequently in the gallon and in the quantities shown in
petroleum industry (e.g., blending crude oil
to produce different octane gasolines), Table 4.13. Product specifications for the
chemical industry (e.g., blending chemicals regular and premium gasolines restrict the
to produce fertilizers and weed killers), and amounts of each component that can be
food industry (e.g., blending ingredients to used in each gasoline product. Table 4.14
produce soft drinks and soups). In this lists the product specifications. Current
section we illustrate how to apply linear commitments to distributors require Grand
programming to a blending problem in the Strand to produce at least 10,000 gallons of
petroleum industry. regular gasoline. The Grand Strand
blending problem is to determine how many
gallons of each component should be used
in the regular gasoline blend and how many
should be used in the premium gasoline
blend. The optimal blending solution should
maximize the firm’s profit, subject to the
constraints on the available petroleum
supplies shown in Table 4.13, the prod-
uct specifications shown in Table 4.14, and
the required 10,000 gallons of regular
gasoline.
The total gallons of each petroleum
component are computed in a similar
fashion.
The total number of gallons of each type of When we combine terms, the objective
gasoline produced is the sum of the number function becomes
of gallons produced using each of the three
petroleum components. Max 0.40x1r + 0.30x2r + 0.06x3r + 0.50x1p
+ 0.40x2p + 0.16x3p
Total Gallons Produced
The limitations on the availability of the
Regular gasoline = x1r + x2r + x3r three petroleum components are
Premium gasoline = x1p + x2p + x3p
x1r + x1p ≤ 5,000 component 1
x2r + x2p ≤ 10,000 component 2
x3r + x3p ≤ 10,000 component 3
Six constraints are now required to meet the The complete linear programming model
product specifications stated in Table 4.14. with six decision variables and 10
constraints is:
The first specification states that component Max 0.40x1r + 0.30x2r + 0.06x3r +
1 can account for no more than 30% of the 0.50x1p + 0.40x2p + 0.16x3p
total gallons of regular gasoline produced.
That is, s.t.
x1r + x1p ≤ 5,000
x1r ≤ 0.30(x1r + x2r + x3r) x2r + x2p ≤ 10,000
x3r + x3p ≤ 10,000
The second product specification listed in x1r ≤ 0.30(x1r + x2r + x3r)
Table 4.14 becomes x2r ≥ 0.40(x1r + x2r + x3r)
x3r ≤ 0.20(x1r + x2r + x3r)
x2r ≥ 0.40(x1r + x2r + x3r) x1p ≥ 0.25(x1p + x2p + x3p)
x2p ≤ 0.45(x1p + x2p + x3p)
Similarly, we write the four remaining x3p ≥ 0.30(x1p + x2p + x3p)
blending specifications listed in Table 4.14 x1r + x2r + x2r ≥ 10,000
as x1r, x2r, x3r, x1p, x2p, x3p ≥ 0