CMNIIACA: Cloud model and niche ideology-based improved ant colony algorithm

In order to improve the convergence properties of basic ant colony algorithm (ACA), a novel type of Cloud Model and Niche Ideology-based Improved Ant Colony Algorithm (CMNIIACA) for solving complex combinational optimization problems is proposed in this paper. Cloud model theory is a powerful tool to convert numerical quantitative analysis to conceptual qualitative analysis. On the basis of introduction of ACA and cloud model theory, a novel qualitative strategy for improving the global optimization properties by use of cloud models is presented in this paper. Then, in order to avoid the stagnation, and to avoid local minima, pseudo-random-proportional action choice rule and elitist preservation strategy are adopted. As the inferior solutions seriously interfered with the searching quality, and inspired by the idea of ecological niche, all the discrete nodes are divided into several groups, the moving scope of each ant agent is then limited. Furthermore, we also limit the trail amount in a maximum-minimum interval. The simulation experiments on CHC150TSP have been performed. The computational results show that the proposed CMNIIACA can effectively improve the global convergence and the evolutional speed of ACA, and the stability of algorithm is also improved effectively. It is obvious that the CMNIIACA presented in this paper is efficient and outperforms ACA.