Binary-Coding-Based Ant Colony Optimization and Its Convergence

Ant colony optimization (ACO for short) is ameta-heuristics for hard combinatorial optimization problems. It is apopulation-based approach that uses exploitation of positive feedbackas well as greedy search.In this paper, genetic algorithm's (GA for short) ideas areintroduced into ACO to present a new binary-coding based ant colony optimization.Compared with the typical ACO, the algorithm is intended to replace theproblem's parameter-space with coding-space, which links ACOwith GA so that the fruits of GA can be applied to ACO directly.Furthermore, it can not only solve general combinatorial optimizationproblems, but also other problems such as function optimization. Based onthe algorithm, it is proved that if the pheromone remainder factor rho isunder the condition of rho>=1, the algorithm can promise to convergeat the optimal, whereas if 0