Introduction

 Most of the optimization problems considered to this point

have had a single objective.
 Often, more than one objective can be identified for a
given problem.
 Maximize Return or Minimize Risk
 Maximize Profit or Minimize Pollution
 These objectives often conflict with one another.
 How do we deal with such situations?

1
Goal Programming (GP)
 Most LP problems have hard constraints that cannot be
violated...
 There are 1,566 labor hours available.
 There is $850,00 available for projects.
 In some cases, hard constraints are too restrictive...
 You have a maximum price in mind when buying a car (this is
your “goal” or target price).
 If you can’t buy the car for this price you’ll likely find a way to
spend more.
 We use soft constraints to represent such goals or targets
we’d like to achieve.
2
A Goal Programming Example:
Hotel Expansion
 We want to expand the convention center in an hotel.
 The types of conference rooms being considered are:

Size (m2) Unit Cost

Small 400 $18,000
Medium 750 $33,000
Large 1,050 $45,150
 We would like to add 5 small, 10 medium and 15 large
conference rooms.
 We also want the total expansion to be 25,000 square
metres and to limit the cost to $1,000,000.
3
Defining the Goals
 Goal 1: The expansion should include approximately
5 small conference rooms.
 Goal 2: The expansion should include approximately
10 medium conference rooms.
 Goal 3: The expansion should include approximately
15 large conference rooms.
 Goal 4: The expansion should consist of
approximately 25,000 square metres.
 Goal 5: The expansion should cost approximately
$1,000,000.
4
Defining the Goal Constraints
 Small Rooms
 
X1  d  d  5
1 1

 Medium Rooms
 
X 2  d  d  10
2 2

 Large Rooms
 
X 3  d  d  15
3 3
where
 
d ,dd , d 0 0

i

i i i

5
Defining the Goal Constraints
 Total Expansion
 
400X 1  750 X 2  1050 X 3  d  d  25000
4 4
 Total Cost (in $1000s)
 
18X 1  33 X 2  45 . 15 X 3  d  d  1000
5 5

where
 
d ,d  0
i i

6
Objective Function

7
Objective Function: Adding weights

8
Objective Function: Standardizing

1   1 1
MIN: (w1 d1  w1 d1 )  (w 2 d 2  w 2 d 2 )  (w 3 d3  w 3 d3 )
     

5 10 15
1 1
 (w 4 d 4  w 4 d 4 )  (w 5 d5  w 5 d5 )
25, 000 1, 000, 000

9
GP Objective Functions
 There are numerous objective functions we could
formulate for a GP problem.
 Minimize the sum of the deviations:

min  ( d  d )
i
-
i


 Problem: The deviations measure different things, so what

does this objective represent?

10
GP Objective Functions (cont’d)
 Minimize the sum of percentage deviations:
1 -
min  ( d i  d i )


i ti

where ti represents the target value of goal i

 Problem: Suppose the first goal is underachieved by 1

small room and the fifth goal is overachieved by $20,000.
 We underachieve goal 1 by 1/5 = 20%
 We overachieve goal 5 by 20,000/1,000,000 = 2%
 This implies being $20,000 over budget is just as undesirable as
having one too few small rooms. Is this true? Only
the decision maker
11 can say for sure.
GP Objective Functions (cont’d)
 Weights can be used in the previous objectives to allow
the decision maker to indicate:
 Desirable vs. undesirable deviations
 The relative importance of various goals
 Minimize the weighted sum of deviations:
min  ( d w  d w )
i
-
i

i

i


i
 Minimize the weighted sum of % deviations:
1 - 
min  ( d i wi  d i wi )
 

i ti

12
Objective Function

1   1 1
MIN: (w1 d1  w1 d1 )  (w 2 d 2  w 2 d 2 )  (w 3 d3  w 3 d3 )
     

5 10 15
1 1
 (w 4 d 4  w 4 d 4 )  (w 5 d5  w 5 d5 )
25, 000 1, 000, 000

13
Defining the Objective
 Assume
 It is undesirable to underachieve any of the first three room
goals (no penalty for overachieving)
 It is undesirable to overachieve or underachieve the 25,000 sq
m expansion goal
 It is undesirable to overachieve the $1,000,000 total cost goal

w 1  w 2  w 3  w 4 w 
w 
MIN: d1  d2  d3  d 4  4
d 4  5
d 5
5 10 15 25, 000 25, 000 1, 000, 000

Initially, we will assume all the above weights equal 1.

14
Comments About GP
 GP involves making trade-offs among the goals until the
most satisfying solution is found.
 GP objective function values should not be compared
because the weights are changed in each iteration.
Compare the solutions!
 An arbitrarily large weight will effectively change a soft
constraint to a hard constraint.
 Hard constraints can be placed on deviational variables.

15
Defining the Goals
 Goal 1: Build approximately 5 small conference rooms.
X1  d1  d1  5
 Goal 2: Build approximately 10 medium conference rooms.

X 2  d 2  d 2  10
 Goal 4: Our target is not to go under 25,000 square meters (our
area demand is 25,000 square meters).

400X1  750X 2  1050X3  d 4  d 4  25000

 Goal 5: Our target is not to go over $1,000,000. (we want to try
not to go over budget, it is ok to go under).
18X1  33X 2  45.15X3  d5  d5  1000
16
 1 1 1
MIN: (w1 d1  w1 d1 )  (w 2 d 2  w 2 d 2 )  (w 3 d3  w 3 d3 )
5 10 15
1     1
 (w 4 d 4  w 4 d 4 )  (w 5 d5  w 5 d5 )
25, 000 1, 000, 000

Goal 4: Our target is not to go under 25,000 square meters

(our area demand is 25,000 square meters).

w 4  0

Goal 5: Our target is not to go over $1,000,000. (we want to

try not to go over budget, it is ok to go under).

w 5  0
17
Summary of Goal Programming
1. Identify the decision variables in the problem.
2. Identify any hard constraints in the problem and formulate them in
the usual way.
3. State the goals of the problem along with their target values.
4. Create constraints using the decision variables that would achieve
the goals exactly.
5. Transform the above constraints into goal constraints by including
deviational variables.
6. Determine which deviational variables represent undesirable
deviations from the goals.
7. Formulate an objective that penalizes the undesirable deviations.
8. Identify appropriate weights for the objective.
9. Solve the problem.
10. Inspect the solution to the problem. If the solution is unacceptable,
return to step 8 and revise the weights as needed.
18
Multiple Objective Linear Programming
(MOLP)
 An MOLP problem is an LP problem with more than one
objective function.
 MOLP problems can be viewed as special types of GP
problems where we must also determine target values for
each goal or objective.
 Analyzing these problems effectively also requires that we
use the MiniMax objective described earlier.

19
An MOLP Example:
The Blackstone Mining Company
 Blackstone Mining runs 2 coal mines in Poland.
 Monthly production by a shift of workers at each mine is
summarized as follows:
Type of Coal Wythe Mine Giles Mine
High-grade 12 tons 4 tons
Medium-grade 4 tons 4 tons
Low-grade 10 tons 20 tons
Cost per month $40,000 $32,000
Gallons of toxic water produced 800 1,250
Life-threatening accidents 0.20 0.45
 Blackstone needs to produce 48 more tons high-grade, 28
more medium-grade, and 100 more low-grade coal.
20
Defining the Decision Variables
 X1 = number of months to schedule an extra shift at the
Wythe county mine
 X2 = number of months to schedule an extra shift at the
Giles county mine

 There are three objectives:

 Min: $40 X1 + $32 X2 } Production costs
 Min: 800 X1 + 1250 X2 } Toxic water
 Min: 0.20 X1 + 0.45 X2 } Accidents

21
Defining the Constraints
 High-grade coal required
 12 X1 + 4 X2 >= 48
 Medium-grade coal required
 4 X1 + 4 X2 >= 28
 Low-grade coal required
 10 X1 + 20 X2 >= 100
 Nonnegativity conditions
 X1, X2 >= 0

22
Handling Multiple Objectives
 If the objectives had target values we could treat them like
the following goals:
 Goal 1: The total cost of productions cost should be
approximately t1.
 Goal 2: The amount of toxic water produce should be
approximately t2.
 Goal 3: The number of life-threatening accidents
should be approximately t3.
 We can solve 3 separate LP problems, independently
optimizing each objective, to find values for t 1, t2 and t3.
23
Summarizing the Solutions
X2
12

11
Feasible Region
10
9
8
7
6 Solution 1
5 (minimum production cost)
Solution 2
4
(minimum toxic water)
3
2
Solution 3
1
(minimum accidents)
0
0 1 2 3 4 5 6 7 8 9 10 11 12 X1
Solution X1 X2 Cost Toxic Water Accidents
1 2.5 4.5 $244 7,625 2.53
2 4.0 3.0 $256 6,950 2.15
3 10.0 0.0 $400 8,000 2.00
24
Defining the Goals
 Goal 1: The total cost of productions cost should
be approximately $244.
 Goal 2: The gallons of toxic water produce should
be approximately 6,950.
 Goal 3: The number of life-threatening accidents
should be approximately 2.0.

25
Defining an Objective
 We can minimize the sum of % deviations as follows:

 40 X 1  32 X 2   244   800 X 1  1250 X 2   6950   0.20 X 1  0.45X 2   2 

MIN: w 1    w2
 
  w3
 


 244   6950   2 

 It can be shown that this is just a linear combination of

the decision variables.
 As a result, this objective will only generate solutions at
corner points of the feasible region (no matter what
weights are used).

26
Defining a Better Objective
MIN: Q
Subject to the additional constraints:

 40X 1  32 X 2   244 
w 1   Q

 244 
 800 X 1  1250 X 2   6950 
w 2   Q

 6950 
 0.20 X 1  0.45X 2   2 
w 3   Q

 2 

 This objective will allow the decision maker to explore non-

corner point solutions of the feasible region.
27
Possible MiniMax Solutions
X2
12

11
Feasible Region
10

6
w1=10, w2=1, w3=1, x1=3.08, x2=3.92
5

4
w1=1, w2=10, w3=1, x1=4.23, x2=2.88
3

1
w1=1, w2=1, w3=10, x1=7.14, x2=1.43
0
0 1 2 3 4 5 6 7 8 9 10 11 12
X1
28
Comments About MOLP
 Solutions obtained using the MiniMax objective are Pareto
Optimal.
 Deviational variables and the MiniMax objective are also
useful in a variety of situations not involving MOLP or GP.
 For minimization objectives the percentage deviation is:
(actual - target)/target
 For maximization objectives the percentage deviation is:
(target - actual)/target
 If a target value is zero, use the weighted deviations rather
than weighted % deviations.
29
 Summary of MOLP
1. Identify the decision variables in the problem.

2. Identify the objectives in the problem and formulate them as usual.

3. Identify the constraints in the problem and formulate them as usual.

4. Solve the problem once for each of the objectives identified in step 2 to
determine the optimal value of each objective.

5. Restate the objectives as goals using the optimal objective values identified
in step 4 as the target values.

6. For each goal, create a deviation function that measures the amount by
which any given solution fails to meet the goal (either as an absolute or a
percentage).

7. For each of the functions identified in step 6, assign a weight to the

function and create a constraint that requires the value of the weighted
deviation function to be less than the MiniMax variable Q.

8. Solve the resulting problem with the objective of minimizing Q.

9. 30Inspect
the solution to the problem. If the solution is unacceptable, adjust
the weights in step 7 and return to step 8.