XI''AN UNIVERSTIY OF ARCHITECTURE AND TECHNOLOGY
The National Natural Science Foundation of China (General Program, Key Program, Major Research Plan)
并行生产线和特定工序生产资源共享模式可以显著改善客户满意度和节约成本. 针对预制构件并行生产线资源配置与生产调度集成优化问题, 基于分解策略和交替迭代优化思想, 提出一种交替式混合果蝇-禁忌搜 索算法(AHFOA_TS) 以最小化拖期惩罚费用. 首先, 通过快速启发式方法产生一较好初始解? 随后固定资源配置方案, 为提高算法局部搜索能力, 通过集成多种局部搜索方式, 设计一种离散果蝇优化算法优化订单指派及调度 方案? 然后固定订单指派及调度方案, 为减少无效搜索次数, 设计一种基于双层变异算子和精英劣解交叉策略的混合禁忌搜索算法以优化资源配置方案. 如此两个阶段交替运行直到满足终止条件. 此外, 还设计了四种基于交 替搜索框架的智能优化算法用于比较. 计算结果表明AHFOA_TS 算法能更有效求解预制构件生产线资源配置与生产调度集成优化问题.
Customer satisfaction and production cost can be improved significantly by parallel production and resources sharing mode. To deal with integrated resources configuration and production scheduling problem for parallel precast production lines, based on decomposition strategy and alternative iterative optimization framework, an alternative hybrid fruit fly and tabu search algorithm (AHFOA_TS) is proposed to minimize total weighted tardiness. In the algorithm, a good initial solution is obtained by a fast heuristic method at first. Then, based on the characteristics of the problem, to enhance the local search ability, an effective discrete fruit fly optimization algorithm is designed to optimize order assignment and scheduling by integrating multiple local search methods. Subsequently, to reduce the number of invalid searches, a hybrid tabu search algorithm based on double-mutation operator and elite inferior solution crossover strategy is designed to optimize the allocation scheme of limited production resources in specific processes. The previous two phases run alternately until the termination criterion is met. For comparison, four hybrid intelligent algorithms are designed based on the alternative search framework. The calculation results verify that, AHFOA_TS performs better than the other four algorithms for dealing with the proposed integrated optimization problem.