IE 607 Heuristic Optimization Ant Colony Optimization 1

IE 607 Heuristic Optimization Ant Colony Optimization 1

Double Bridge Experiment 2

Behavior of Real Ants Find the Shortest Path to Food Resource Pheromone Is Laid by Ants along the Trail Pheromone Evaporates over Time Pheromone Intensity Increases with Number of Ants Using Trail Good Paths Are Reinforced And Bad Paths Gradually Disappear 3

ACO Meta-Heuristic Optimization Method Inspired by Real Ants First published by Marco Dorigo (1992) as his dissertation Is currently greatly expanding in applications and interest, mainly centered in Europe Positive & Negative Feedback Constructive Greedy Heuristic Population-based Method 4

Applicatio n TSP QAP VRP Telecommunication Network Scheduling Graph Coloring Water Distribution Network etc 5

Methodolo gy ACO Algorithm ACO Set all parameters and initialize the pheromone trails Loop Sub-Loop Construct solutions based on the state transition rule Apply the online pheromone update rule Continue until all ants have been generated Apply Local Search Evaluate all solutions and record the best solution so far 6

Methodolo gy Overview of ACO Algorithm Each ant represents a complete solution Online updating is performed each time after an ant constructed a solution: more chance to exploration Local search is applied after all ants construct solutions Offline updating is employed after the local search: allow good ants to contribute 7

Methodolo gy Parameters of ACO Algorithm : Pheromone trail of combination (i, j) : Local heuristic of combination (i, j) : Transition probability of combination (i, j) : Relative importance of pheromone trail : Relative importance of local heuristic : Determines the relative importance of exploitation versus exploration 8

Methodolo gy Ant System (AS) – the earliest version of ACO State Transition Probability Pheromone Update Rule 9

Methodolo gy ASelite ASrank 10

Methodolo gy Ant-Q & Ant Colony System (ACS) Exploitation Exploration Local Updating (Online Updating) Global Updating (Offline Updating) 11

Methodolo gy Max-Min Ant System (MMAS) ANTS 12

Website & Books http: //iridia. ulb. ac. be/~mdorigo/ACO/AC O. html Bonabeau E. , M. Dorigo & T. Theraulaz (1999). From Natural to Artificial Swarm Intelligence. New York: Oxford University Press. Corne D. , M. Dorigo & F. Glover, Editors (1999). New Ideas in Optimisation. Mc. Graw-Hill. 13