Publikationen ohne Studienabschluss

Complex mechatronic products are usually decomposed into several sub-systems for their development. These sub-systems are developed in parallel or even independently based on their specifications and use cases. The application of model-based solution patterns is an effective way to comprehensively and efficiently describe the available knowledge about the sub-systems. This contribution proposes an approach to support the selection and application of model-based solution patterns. The approach, based on a metamodel for solution patterns using SysML, describes the process for selecting solution patterns and aligning requirements and constraints with the as-is properties of the sub-systems. Additionally, the approach supports the design of solution patterns taking into account special knowledge from the development of the sub-systems as well as the usage of the solution patterns in different systems and contexts. As an example, an application scenario of a specific load cell within a measurement system is explained.

Operating current advanced production systems, including Cyber-Physical Systems, often requires profound programming skills and configuration knowledge, creating a disconnect between human cognition and system operations. To address this, we suggest developing cognitive algorithms that can simulate and anticipate teaming partners' cognitive processes, enhancing and smoothing collaboration in problem-solving processes. Our proposed solution entails creating a cognitive system that minimizes human cognitive load and stress by developing models reflecting humans individual problem-solving capabilities and potential cognitive states. Further, we aim to devise algorithms that simulate individual decision processes and virtual bargaining procedures that anticipate actions, adjusting the system’s behavior towards efficient goal-oriented outcomes. Future steps include the development of benchmark sets tailored for specific use cases and human-system interactions. We plan to refine and test algorithms for detecting and inferring cognitive states of human partners. This process requires incorporating theoretical approaches and adapting existing algorithms to simulate and predict human cognitive processes of problem-solving with regards to cognitive states. The objective is to develop cognitive and computational models that enable production systems to become equal team members alongside humans in diverse scenarios, paving the way for more efficient, effective goal-oriented solutions.

Robust Design offers a coherent and widely appreciated approach for the parametric exploration of the design space by means of simulation or experimentation, which is well-established in the quality-by design domain. From the perspective of design research, however, this only addresses a relatively narrow part of the design process and is not fully integrated with other design decisions such as concept exploration, the suitable configuration of system elements, or the design of interfaces. Particularly in light of the growing importance of developing technologically advanced and “smart” systems, it seems that a new methodical perspective on Robust Design is needed. Against this background, this paper consolidates knowledge and insights from different research fields and industry sectors. On this basis, new angles to the discussion on product robustness in different domains are explored in order to suggests directions for action and new research areas, both with respect to a methodical RD approach as well as the question of systematic research procedures.

Digital Twins are being used more and more frequently and provide information from the Real Twin for different applications. Measurements on the Real Twin are required to obtain information, which in many cases requires the installation of supplementary sensors. For their conception and design, it is particularly important that the measuring principles are selected purposefully and the appropriate sensors are integrated at the goal-oriented measuring positions without impairing the functions and other properties of the Real Twin by the integration of these sensors. In this article, a "Design for Digital Twin" approach is discussed for the systematic procedure and demonstrated using a multi-staged gearbox as a concrete example. The approach focuses on the mechanical and hardware side of the Real Twin. For the systematic conception and design of the Digital Twin solution, an understanding of the stakeholder demands and the expected use cases is necessary. Based on the stakeholder demands and use cases, the relevant product properties can be determined. Using the relevant properties, an iterative process of conception, design, and analysis takes place. The conception is carried out by means of target-oriented cause-effect analyses, taking into account systemic interrelations of the Real Twin components and systematics for the selection of measurement principles. Systemic considerations, combined with an effect graph, allow for the analysis and evaluation of disturbing factors.

Product development is characterised by numerous synthesis and analysis loops. Analysis provides information on the fulfillment of the required properties of the system under development. Analysis results therefore are an important basis for further synthesis steps. In the context of Model-Based Systems Engineering (MBSE), different types of simulation models play an important role. The overall system (product) can be broken down to system elements. For each system element a set of models has to be established, that provides the required degrees of fidelity, representations of system properties, flows, etc. reflecting the variety of modelling purposes. These models have to be integrated horizontally (same system level along the relevant flows of material, energy, and information) and vertically (aggregation from subsystem level to the overall system level, refinement in the opposite direction) to create a holistic model-based system representation. An important and challenging task is to identify and shape relevant subsystem models. In order to define an appropriate structure of these models, model developers may utilize criteria like selected properties of system elements and interrelations, their degree of detail or modelling assumptions. The relevant criteria have to be made transparent. For this purpose, the paper discusses the concept of model signatures that contain relevant meta information about each single model of all system elements, subsystems up to models of the overall system. This standardized meta information enables an identification and selection of those models and the decision on the necessary model integration. The concept is discussed on the basis of a roller bearing as an example out of an electro-mechanical drivetrain. A potential analysis provides information about the possible usage of the model signatures concept.