An adaptive ant colony optimization was proposed to solve

the assembly line balancing problem(ALBP).According to the characteristics of

the ALBP,a method of solution constructing strategy was developed,and a better differentiation of objective function was proposed to appraisal solution quality.However,general ant algorithm often falls into local optimal and consume excessive time, in order to overcome these shortcoming,an improved ACO was presented by adaptive adjustment of the parameters in the algorithm, which has a good ability of searching better solution at higher convergence speed. Finally, the proposed algorithm was tested and compared against best known algorithms reported in the literatures, and the experimental results indicate the feasibility and effectiveness of the proposed algorithm.

提出了一种自适应蚁群算法,用以求解装配线平衡问题。在该算法中，针对装配线平衡问题的具体特点，设计了一种蚂蚁分配方案可行解的构造策略，提出了一种比传统方法区分度更高的评价解质量的目标函数，同时为了克服蚁群算法易陷入局部最优和收敛速度慢等缺陷，通过自适应地调整算法的挥发度等系数，在保证收敛速度的条件下提高了解的全局性。最后，通过实例验证，证明了算法的可行性和有效性。

装配线平衡, 蚁群算法, 自适应, 人工智能