Digital twin for the single and small batch production

本研究的目的是探讨在制造过程中所捕获的数据之间的依赖关系，以开发一个生产孪生模型。直接和间接地，相互依赖的数据被检查的因果关系，从而形成一个有效的和有效的改进制造过程的基础。数据的使用进一步支持模拟生产过程，并大大缩短上市时间。数据的评估和使用允许单件和小批量生产优化规划，从而缩短交货时间。生产的各个部分的优化是无效的。只有通过优化整个生产系统才能实现效率的提高。然而，这需要了解整个生产过程及其各个子过程的依赖性。对过程和参数的依赖性的理解和准确表示允许预测某些事件，并最终有助于在负面事件的情况下进行干预。对流程进行主动干预的要求是理解和记录各个流程或子流程的因果关系，这可以通过理解数据依赖性来实现。识别出的数据以及这些数据之间的因果关系最终被组合成数字生产孪生模型。要取得的成果可分为七个次级目标：a）在单个和小批量生产公司中创建所有生成的和相关的数据的通用目录B）在单个和小批量生产公司中创建所有存储的数据的通用目录c）开发数据结构之间的依赖模式d）以多维矩阵的形式描绘数据结构中的依赖关系e）识别积极和消极的数据依赖关系f）识别不同数据结构的因果关系形式的依赖关系g）将数据和因果关系整合到数字生产孪生模型中。

The goal of this research project is to investigate the dependencies among the captured data in manufacturing processes in order to develop a production twin model. Both, directly and indirectly, interdependent data is examined for cause-effect relationships, and thus form the basis for an efficient and effective improvement of manufacturing processes. The use of data further supports to simulate production processes and to shorten the time-to-market significantly. An evaluation and use of data allows the single and small series production to optimize the planning whereby lead times can be reduced. The optimization of individual sections of the production is not effective. An increase of the efficiency can only be achieved by optimizing the overall production system. However, this requires an understanding of the entire production process and dependencies of their individual sub-processes. An understanding and accurate representation of the dependency of processes and parameters allow then to predict certain events and finally help to interfere in case of negative events. The requirements for a proactive intervention in the process is the understanding and documentation of cause-effect relationships of individual processes or sub-processes, which can be attained through the understanding of data dependencies. The identified data and cause-effect relationships between these data are combined finally to a digital production twin model. The results to be achieved can be divided into seven sub-goals: a) Creation of a generic catalog of all generated and relevant data in single and small series production companies b) Creation of a generic catalog of all stored data in single and small series production companies c) Development of dependency patterns between data structures d) Portrait of the dependencies in data structures in the form of a multidimensional matrix e) Identification of positive and negative data dependencies f) Identification of dependencies in the form of causeeffect relationships of different data structures g) Integration of the data and the cause-effect relationships to a digital production twin model.