中文摘要：

针对带多处理器的混合流水车间调度问题（hybrid flow shop scheduling with multiprocessor task problems），以最小化所有工件的最大完成时间（makespan）为优化目标，提出一种融合了改进的人工鱼群算法和禁忌搜索算法的混合算法。首先改进人工鱼群算法相关行为及实验优选算法参数，提高了人工鱼群算法收敛速度和精度；然后结合人工鱼群算法收敛快和禁忌算法局部搜索能力强的特点，利用改进的人工鱼群算法进行全局搜索，获得较好的优化解域，再通过禁忌算法在优化解域内进行局部寻优，得到一个最终满意的优化解。基于180个标准算例，算法实验结果表明混合算法的优化性能明显优于禁忌算法和粒子群算法，并且很接近改进的遗传算法。

英文摘要：

For the hybrid flow shop scheduling problem with multiprocessor tasks, a hybrid algorithm was proposed to minimize the maximum completion time （makespan ） of all jobs, which is based on an improved artificial fish swarm algorithm and a tabu search algorithm Firstly, the convergence speed and accuracy of an artificial fish swarm algorithm was improved by improving the related behavior of an artificial fish and optimizing algorithm parameters. Secondly, according that the artificial fish algorithm converges fast and local search ability of tabu algorithm is powegrul, the improved artificial fish swarm algorithm was used for global search to fast get better optimization solution set, in which then the tabu search algorithm was used to seek approximate solution. The computational results based on 180 well-known instances demonstrate the effectiveness of the proposed hybrid algorithm has better optimization results than tabu search algorithm and particle swarm optimization algorithm, and is very close to the improved genetic algorithm.