DEEP - Data and knowledge-based decision support in the early development of robust products

• Applicants: Professor Dr.-Ing. Sven Matthiesen; Professor Dr.-Ing. Sandro Wartzack
• Subject Area: Engineering Design, Machine Elements, Product Development
• Term: since 2021
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 467789897

In today's dynamic market, reducing costly iterations is crucial for competitiveness of a company. Various deviations can reduce the functional fulfillment of a product, result in unwanted quality losses. This is where aspects of Robust Design (RD) gain importance by focusing on reducing the sensitivity to such deviations. Traditional RD approaches typically focus on later stages of product development when the detailed geometry of the product is available. However, conceptual design in the early stages of development has a significant impact on the overall robustness, performance, and cost of a product. Early robustness considerations can avoid costly design changes and high tolerance requirements. Despite the awareness of the enormous importance of early robustness assessment, this aspect often remains unused in industrial practice. Early RD methods are difficult to apply to specific design tasks due to their high level of abstraction. They are often based on expert-based RD principles, which are difficult to apply due to their large number and partial inconsistency. Therefore, a robustness evaluation of different product concepts remains challenging. In order to comprehensively evaluate the robustness to deviations in the early phases, both the system design and the geometric design of the product concept must be evaluated. The Embodiment Function Relation and Tolerance (EFRT) model has been developed for this purpose. The EFRT model allows for a targeted investigation of the robustness of product concepts against deviations. It therefore offers the potential to be used for model-based statements to reduce subjective assessments of robustness. Ultimately, design decisions should also be made for the synthesis of robust products, for which the versatility of the model-based statements is a prerequisite. Although the early applicability of robustness evaluation with the EFRT model offers promising approaches to support the design decision, the potential to increase the versatility of these evaluations remains unexplored. The problem is that the limited robustness of robustness evaluation results in subjective and error-prone design decisions, which can lead to insufficient product robustness and associated increased costs. Therefore, the main goal of this research project is to increase the versatility of robustness evaluation by developing an approach to support design decisions. The approach will overcome the two main barriers of (1) insufficient modeling depth and (2) subjective assessment by design engineers.

• DFG Programme: Research Grants