Assignment Problems

Guoming Tang
 A Practical Problem
Workers: A, B, C, D Jobs: P, Q, R, S.
Cost matrix: Job P Job Q Job R Job S
Worker A 1 2 3 4
Worker B 2 4 6 8
Worker C 3 6 9 12
Worker D 4 8 12 16

Given: each worker need perform only one job and

each job need be assigned to only one worker.
Question: how to assign the jobs to the workers
to minimize the cost?
2012-11-21 CSC 545 - Graduate Lecture 2
 Assignment Problem
Teachers vs. Courses
Performance
Maximize the total performance
Jobs vs. Machines
Time
Minimize the total time

2012-11-21 CSC 545 - Graduate Lecture 3

 Formulation of Assignment Problem
Consider m workers to whom n jobs are assigned.
The cost of assigning worker i to job j is cij.

Let 1 , if job j is assigned to worker i

xij =
0 , if job j is not assigned to worker i

2012-11-21 CSC 545 - Graduate Lecture 4

 Formulation of Assignment Problem
Objective function:

Constraints:

Each worker is assigned one job

Each job is assigned one worker

2012-11-21 CSC 545 - Graduate Lecture 5
 Solution of Assignment Problem
Brute-force method
Enumerate all candidates sets
n! possible assignment sets
Exponential runtime complexity

4! = 24 10! = 3,628,800 100! 9.3 10157

2012-11-21 CSC 545 - Graduate Lecture 6

 Solution of Assignment Problem
Simplex method
Is it feasible to solve AP? Yes.
[The Integrality Theorem] If a transshipment
problem: minimize cx subject to Ax=b, x0, such
that all the components of b are integers, has at
least one feasible solution, then it has an integer-
valued feasible solution; if it has an optimal
solution, then it has an integer-valued optimal
solution.

2012-11-21 CSC 545 - Graduate Lecture 7

 Solution of Assignment Problem
Simplex method
More variables (an n assignments needs n2
variables.)
More slack variables result in a sparse dictionary
matrix, which may lead to more iterations.
Inefficient.

2012-11-21 CSC 545 - Graduate Lecture 8

 Solution of Assignment Problem
Network simplex method
Tree based network optimization method
Can apply to transshipment problem, maximum
flows through networks
Works well in practice for assignment problems.
Is there any easier way to solve the assignment
problem?

2012-11-21 CSC 545 - Graduate Lecture 9

 Hungarian Method
Introduction

A combinatorial optimization algorithm

Based on two Hungarian mathematicians works
Developed and published by Harold Kuhn, 1955
Polynomial runtime complexity
[Wikipedia] O(n4), can be modified to O(n3)
Much easier to implement

2012-11-21 CSC 545 - Graduate Lecture 10

 Hungarian Method
Process (1/5)
Assume the cost matrix
5 6 7 6 0 0 2 1
4 3 2 3 2 0 0 1
c= c =
2 3 5 2 0 0 3 0
5 5 2 8 3 2 0 6
For each row, subtract the minimum number in that
row from all numbers in that row; do the same for each
column.
[Network optimization: Theorem 2.9] There exists an
optimal solution, when # of assignments = minimum #
of lines required to cover all 0s.
2012-11-21 CSC 545 - Graduate Lecture 11
 Hungarian Method
Process(2/5)
How can we find the optimal solution?
One efficient way in [Network optimization]
(a) Locate a row/column in c with exactly one 0, circle it
and draw a vertical/horizontal line through it. If no such
row/column exists, locate a row/column with the smallest
number of 0s.
(b) Repeat (a) till every 0 in the matrix has at least one line
through it.
(c) If # of lines equal to n, the circled 0s show the optimal
solution.
Apply it to c.

2012-11-21 CSC 545 - Graduate Lecture 12

 Hungarian Method
Process(3/5)
Why is it optimal? 0 0 2 1
2 0 0 1
c =
0 0 3 0
3 2 0 6

[Network optimization: Theorem 2.17]

Suppose c is the cost matrix and c is the matrix
obtained by adding a number t to each element in the
ith row or to each element in the ith column. Then a
solution is optimal with respect to c if and only if it is
optimal with respect to c.

2012-11-21 CSC 545 - Graduate Lecture 13

 Hungarian Method
Process(4/5)
Is that all? No. 3 0 2 1
2 4 0 1
c =
0 0 3 0
3 2 0 6

No matching can result in 0 total cost.

Remember: (c) If # of lines equals to n, we find the
optimal solution.
Then what if # of lines is not equal to n?

2012-11-21 CSC 545 - Graduate Lecture 14

 Hungarian Method
Process(5/5)
3 0 2 1
Modify c further:
2 4 0 1
c =
0 0 3 0
3 2 0 6
Subtract the smallest uncovered element from each
uncovered element and add it to each doubly
covered element.
2 0 2 0
1 4 0 0
Check the solution. c=
0 1 4 0
2 2 0 5
2012-11-21 CSC 545 - Graduate Lecture 15
 Hungarian Method
the Whole Course
1. Given the cost matrix c (nn), get modified c:
(a) For each row, subtract the minimum number in that row from all
numbers in that row
(b) Do the same for each column.
2. Check if there exists an optimal solution:
(a) Locate a row/column in modified matrix with exactly one 0, circle it
and draw a vertical/horizontal line through it. If no such row/column
exists, locate a row/column with the smallest number of 0s.
(b) Repeat (a) until every 0 in the matrix has at least one line through it.
(c) If # of lines = n, the circled 0s show the optimal solution, end; if not,
go to step 3.
3. Further modify c:
(a) Subtract the smallest uncovered element from each uncovered
element and add it to each doubly covered element.
(b) Repeat step 2.

2012-11-21 CSC 545 - Graduate Lecture 16

 Hungarian Method
Special Considerations
Special considerations: Solutions:
Hungarian method Add dummy
requires # of rows = # of rows/columns with 0
columns. What if not? assignment costs
Considering a case like Assign cij = +M
assigning workers to jobs, Create a loss matrix
what if worker i cannot
do job j ?
Maximization objective,
such as maximize the
profits.

2012-11-21 CSC 545 - Graduate Lecture 17

 Hungarian Method
an Example (1/7)
Teachers: A, B, C, D; Courses: P, Q, R
Teachers performance matrix
P Q R
A 90 76 67
B 68 70 69
C 75 72 71
D 69 65 65
Assign courses to teachers to maximize the sums
of their performance.
2012-11-21 CSC 545 - Graduate Lecture 18
 Hungarian Method
an Example (2/7)
Transfer Maximum to Minimum:
Create the loss matrix: subtract each score in each column from the
highest score in that column.
P Q R P Q R S
A 0 0 4 A 0 0 4 0
B 22 6 2 B 22 6 2 0
C 15 4 0 C 15 4 0 0
D 21 11 6 D 21 11 6 0
Cost matrix c
Add a dummy column with 0 score loss.

2012-11-21 CSC 545 - Graduate Lecture 19

 Hungarian Method
an Example (3/7)

Step 1. Given the cost matrix c, get modified c

0 0 4 0 0 0 4 0
22 6 2 0 22 6 2 0
c= c=
15 4 0 0 15 4 0 0
21 11 6 0 21 11 6 0

2012-11-21 CSC 545 - Graduate Lecture 20

 Hungarian Method
an Example (4/7)

Step 2. check if there exists an optimal solution

0 0 4 0
22 6 2 0
c=
15 4 0 0
21 11 6 0

# of lines 4, go to step 3.

2012-11-21 CSC 545 - Graduate Lecture 21

 Hungarian Method
an Example (5/7)

Step 3. Further modify c

0 0 4 0 0 0 8 4
22 6 2 0 18 2 2 0
c= c=
15 4 0 0 11 0 0 0
21 11 6 0 17 7 6 0

# of lines 4, continue step 3.

2012-11-21 CSC 545 - Graduate Lecture 22

 Hungarian Method
an Example (6/7)

Step 3. Further modify c

0 0 8 4 0 0 8 6
18 2 2 0 16 0 0 0
c= c=
11 0 0 0 11 0 0 2
17 7 6 0 15 5 4 0

# of lines = 4, get the optimal solution.

2012-11-21 CSC 545 - Graduate Lecture 23

 Hungarian Method
an Example (7/7)
The optimal solution

0 0 8 6
16 0 0 0
c=
11 0 0 2
15 5 4 0
Assignment:
AP, BR, CQ or AP, BQ, CR
Maximum score:
Z = 90 +69 +72 = 90 +70 +71 = 231

2012-11-21 CSC 545 - Graduate Lecture 24

 Tools for AP & LP
A Parametric Visualization Software for the Assignment
problem. University of Macedonia, Greece.
(http://users.uom.gr/~samaras/gr/yujor/yujor.htm )
LINGO: a comprehensive tool designed to help you build and
solve linear, optimization models quickly, easily, and
efficiently. ( http://www.lindo.com/ )
MATLAB: linprog function for general LP; Toolbox YALMIP for
(mixed) integer LP.

2012-11-21 CSC 545 - Graduate Lecture 25

Thanks for Your Attention.

Assignment Problems
Guoming Tang