A Novel Cloud-Based Fuzzy Self-Adaptive Ant Colony System

An important issue of ant colony optimization (ACO) is how to keep the balance between the exploration in search space regions and the exploitation of the search experience gathered so far. By using a more exploitative pseudo-random-proportional selection rule, ant colony system (ACS) can obtain better results in experiments. But it is limited by finite search space. In this paper, a novel cloud-based fuzzy self-adaptive ant colony system (CFSACS) based on ACS is proposed, in which cloud model is used as the fuzzy membership function and a self-adaptive mechanism is constructed. By using the self-adaptive mechanism and the pheromone updating rule of better solution which is determined by the membership function uncertainly, CFSACS can explore search space more effectively than ACS. The proposed CFSACS is demonstrated to be convergent by analyzing the probability model of it. Moreover, the simulation results show that the CFSACS is more effective than both ACS and MMAS.