随着经济全球化不断深入，制造业逐渐从集中式生产向分布式协作生产发展，多样化的生产环境给生产资源调度带来巨大挑战。本课题拟围绕多条异构柔性阻塞流水线和装配线，考虑不确定生产参数，结合经济、绿色等生产指标，以超启发式算法为基础，研究不确定环境下多目标异构分布式阻塞流水车间调度和装配调度优化问题。通过分析实际生产过程，提炼生产约束和生产目标，建立数学模型；从工件分配、工件排序、机器选择等子问题入手，挖掘问题特定知识，提出有效的构造型启发式算法和优化操作，构建低层领域知识规则库；以群智能优化为基础，实现多种群协同搜索的高层操作管理策略；建立低层与高层通信机制及操作序列的自适应学习机制，提出群智协同的学习型超启发式算法；结合计算机仿真开展模型及方法验证，并探讨其在实际生产中的应用。本课题研究成果可为解决实际生产中同类调度优化问题提供理论基础和关键技术支撑，进一步促进分布式生产调度理论和技术的发展。

With the deepening of economic globalization, the manufacturing industry is gradually developing from the centralized production to the distributed collaborative production, while the diversified production environment brings great challenges for the scheduling of production resource. Surrounding multiple heterogeneous flexible blocking flow lines, considering the uncertainty production parameters and combining the economic and green production target, based on the hyper-heuristic, this project intends to investigate the optimization problems of the multi-objective heterogeneous distributed blocking flow-shop scheduling and assembly scheduling under uncertain environment. Through the analysis of the practical production process, the production constraints and production objectives are extracted, and the mathematical models are established. From the sub problems such as the allocation of jobs, the sorting of jobs and the selection of machines, the problem-specific knowledge is mined. The effective constructive heuristics and optimization operators are proposed, and the low-level rule base with domain knowledge is constructed. Based on the swarm intelligence optimization, the multi-swarm collaboration search strategy is presented as the high-level operation management strategy. The communication mechanism between the low-level and the high-level along with the self-adaptive learning mechanism of the operator sequence are constructed. The learning hyper-heuristics with swarm intelligence collaboration are proposed. The validation of models and methods are conducted by combining the computer simulation, while their application in practical production is discussed. The achievements of this project can provide the theoretical foundation and the support of key technologies for solving the similar scheduling problems in practical production, and further facilitate the development of theory and technology for the distributed production scheduling.