The first ant colony optimization (ACO) called ant system was inspired through studying of the behaviour of ants in 1991 by Macro Dorigo and co-workers. An ant colony is highly organized, in which one interacting with others through pheromone in perfect harmony. Optimization problems can be solved through simulating ant’s behaviours. Since the first ant system algorithm was proposed, there is a lot of development in ACO. In ant colony system algorithm, local pheromone is used for ants to search optimum result. However, high magnitude of computing is its deficiency and sometimes it is inefficient. Thomas Stützle etal. Introduced MAX-MIN Ant System (MMAS) in 2000. It is one of the best algorithms of ACO. It limits total pheromone in every trip or sub-union to avoid local convergence. However, the limitation of pheromone slows down convergence rate in MMAS.

1. OPTIMIZATION BY ANT
COLONY METHOD

2. CONTENT
 defination of optimization
 ACO concept
 ACO system
 ACO system cont.
 ANT foraging
 Implementation
 Applications
 Advantages & Disadvantages
 Sources
 conclusions
 References

3. What is Optimization?
 Procedure to make a system or design as
effective, especially the mathematical
techniques involved. ( Meta-Heuristics)
 Finding Best Solution
 Minimal Cost (Design)
 Minimal Error (Parameter Calibration)
 Maximal Profit (Management)
 Maximal Utility (Economics)

4. 4
ACO Concept
 Ants (blind) navigate from nest to food source
 Shortest path is discovered via pheromone trails
 First ant moves at random
 pheromone is deposited on path
 ants detect lead ant’s path, inclined to follow
 more pheromone on path increases probability of path
being followed

5. 5
ACO System
 Virtual “trail” accumulated on path segments
 Starting node selected at random
 Path selected at random
 based on amount of “trail” present on possible paths
from starting node
 higher probability for paths with more “trail”
 Ant reaches next node, selects next path
 Continues until reaches starting node
 Finished “tour” is a solution

6. 6
ACO System, cont.
 A completed tour is analyzed for optimality
 “Trail” amount adjusted to favor better solutions
 better solutions receive more trail
 worse solutions receive less trail
 higher probability of ant selecting path that is part of a
better-performing tour
 New cycle is performed
 Repeated until most ants select the same tour on
every cycle (convergence to solution)

7. 7

8. 8

9. 9

10. 10

11. 11
Implementation
 Can be used for both Static and Dynamic
Combinatorial optimization problems
 Convergence is guaranteed, although the
speed is unknown
 Value
 Solution

12. Existing Nature-Inspired Algorithms

13. 13
Applications
Efficiently Solves NP hard Problems
 Routing
 TSP (Traveling Salesman Problem)
 Vehicle Routing
 Sequential Ordering
 Assignment
 QAP (Quadratic Assignment Problem)
 Graph Coloring
 Generalized Assignment
 Frequency Assignment
 University Course Time Scheduling
43
5
2
1

14. 14
Applications
 Other
 Shortest Common Sequence
 Constraint Satisfaction
 2D-HP protein folding
 Bin Packing
 Machine Learning
 Classification Rules
 Bayesian networks
 Fuzzy systems
 Network Routing
 Connection oriented network routing
 Connection network routing
 Optical network routing

15. 15
Advantages and Disadvantages,
cont.
 Can be used in dynamic applications (adapts to
changes such as new distances, etc.)
 Has been applied to a wide variety of applications
 As with GAs, good choice for constrained discrete
problems (not a gradient-based algorithm)

16. 16
Sources
 Dorigo, Marco and Stützle, Thomas. (2004) Ant Colony Optimization,
Cambridge, MA: The MIT Press.
 Dorigo, Marco, Gambardella, Luca M., Middendorf, Martin. (2002)
“Guest Editorial,” IEEE Transactions on Evolutionary Computation,
6(4): 317-320.
 Thompson, Jonathan, “Ant Colony Optimization.”
http://www.orsoc.org.uk/region/regional/swords/swords.ppt, accessed
April 24, 2005.
 Camp, Charles V., Bichon, Barron, J. and Stovall, Scott P. (2005)
“Design of Steel Frames Using Ant Colony Optimization,” Journal of
Structural Engineeering, 131 (3):369-379.
 Fjalldal, Johann Bragi, “An Introduction to Ant Colony Algorithms.”
http://www.informatics.sussex.ac.uk/research/nlp/gazdar/teach/atc/199
9/web/johannf/ants.html, accessed April 24, 2005.

17. 17
Advantages and Disadvantages
 For TSPs (Traveling Salesman Problem), relatively efficient
 for a small number of nodes, TSPs can be solved by
exhaustive search
 for a large number of nodes, TSPs are very computationally
difficult to solve (NP-hard) – exponential time to
convergence
 Performs better against other global optimization techniques
for TSP (neural net, genetic algorithms, simulated annealing)
 Compared to GAs (Genetic Algorithms):
 retains memory of entire colony instead of previous
generation only
 less affected by poor initial solutions (due to combination of
random path selection and colony memory)

18. Estimation and simulation, end
users; field work – tracer studies,
pressure tests, case studies;
contaminant and water security –
detection, source identification,
response; network vulnerability –
security assessments, network
reliability,
CONCLUSION