Supporting evolution of automated material flow systems as part of CPPS by using coupled meta models

The software of automated Material Flow Systems (aMFS) is usually tightly coupled to the system in which it is embedded. Reuse of control software of aMFS is commonly achieved through copying, pasting, and modifying existing control code from a similar system. When a module of an aMFS is updated in production, e.g. to resolve issues, the program code and hardware components are changed. The change is usually documented informally or resides only in the mind of the automation engineer. When new hardware is added to a module, the control software often has to be changed in many different parts, or a new control function has to be developed, which wraps the existing code. This way of handling the development and evolution of aMFSs is error-prone and reduces the software quality. A modular, model-based software architecture for aMFS can reduce the error-proneness and improve the software quality. In this paper we propose a model-based development method for aMFS. It is used to create a modular software architecture that describes composable modules. The models, their meta models, and the program code therein are coupled with each other via mappings and consistency rules, so that the consistency can be validated. Bidirectional transformations between the program code and models can be created. As part of the documentation, changes in the program code and model must be synchronized.