作业车间调度算法资源云服务化方法

Servitization of Job Shop Scheduling Algorithms

针对制造执行系统间调度算法资源共享度不高,并且难以高效地从候选调度算法资源中选择合适的求解算法进行调度排产的问题,研究了云制造环境下的作业车间调度算法资源云服务化方法。首先基于OWL-S(ontology web language for services)建立调度算法资源的本体数据模型以及发布规范,然后为了提高算法服务请求者满意度和算法资源提供者经济收益,提出了基于加权多维特征融合的量化匹配与推荐方法。最后,通过搭建原型系统,并利用哈电机的历史数据进行仿真实验。结果表明,所提出的数据模型和推荐算法能够有效的提升调度算法资源的共享程度和选择速度。

The scheduling algorithm resources are not yet fully shared betwwen manufacturing execution system,and it is difficult to efficiently select the appropriate scheduling algorithm for scheduling problem from the candidate scheduling resources. To resolve this issue,the servitization of job shop scheduling algorithm resource in cloud manufacturing was studied. Firstly,an OWL-S-based ontology data model and publication specification of scheduling algorithm resource was given,and then a quantitative matching and recommending method based on weighted multidimensional feature fusion was presented,in order to improve the satisfaction of algorithm service requesters and the economic benefits of algorithm resource providers. Finally,through building a prototype system and using the historical data of Harbin Electrical Machinery Plant as a verification case,the results proved that the data model and recommend algorithm can effectively promote the sharing degree and selection speed of scheduling resources.