Abstract: In view of Distributed Design Resources Dynamic Scheduling(DDRDS), a varying-space activated based scheduling mechanism was proposed. Concepts of trust degree, reliability, load rate and complexity were introduced, and varying-space activated task option principles were defined by using multi-attribute decision making based on Ordered Weighted Average(OWA)operator. Through Discrete Particle Swarm Optimization(DPSO), both mathematic description of DDRDS and particle fitness calculation method were given. The local update varying-space shift strategy was designed to prevent disturbance event causing scheduling failure, and the rapid solution of DDRDS sequence was realized through particle evolution and varying-space shift. Thus the reuse efficiency and accuracy of design resources were improved. A CNC machine tool's scheme design, overall precision design and stiffness design resource scheduling were taken as examples to validate the effectiveness and feasibility of proposed method.

Key words: dynamic scheduling, varying-space activated, discrete particle swarm, multi-attribute decision, numerical control machine tool

摘要： 针对分布式环境下设计资源动态调度问题，提出一种基于变区域激活的任务调度机制。引入信任度、可靠性、负载率、复杂度等概念，运用基于有序加权平均算子的多属性决策方法定义变区域激活任务选入原则。采用离散粒子群算法，给出分布式设计资源动态调度问题的数学描述和粒子适应度计算方法，为防止扰动事件引起调度失败，设计了局部更新变区域平移策略，通过粒子进化和变区域平移，实现分布式设计资源动态调度序列的快速求解，进而提高了设计资源的重用效率和准确性。以数控机床整机方案设计、整机精度、刚度设计资源调度为例，对上述方法的有效性和可行性进行了验证。

关键词: 动态调度, 变区域激活, 离散粒子群, 多属性决策