Simulation Model Selection Process Using Complexity Measurement

Regarding the requirement of continuous development, and adaptation to new changes in the industrial automation field, systems are becoming progressively more complex. In order to cope with the increasing complexity, systems are developed model-based using libraries of predefined models. Various modeling processes assist in simplifying the understanding of the systems deployed in heterogeneous and dynamic environments. On the other hand, the models are growing in complexity to represent the assets precisely. The complexity affects various attributes of the models such as understandability, ease of use, scalability, performance, testing, and verification. Therefore, one of the main challenges of analyzing simulation models for different use cases in different development life cycle phases is complexity assessment. A comprehensive complexity analysis approach enables achieving the proper selection of the models among all available alternatives. This paper presents several approaches for models' complexity assessment from the literature and proposes an approach to select the appropriate simulation models among all alternatives. This approach is realized with a framework to depict the process of how to achieve the required models according to the complexity value and the estimated complexity budget for various use cases. To validate the proposed approach, representative industry models of a Robot Vacuum Gripper are used to select the model based on the complexity metric in simulation.