Optimization-driven design offers advantages over traditional experience-based mechanical design. As an example, topology optimization can be a powerful tool to generate body shapes for Additive Manufacturing (AM). This is helpful, when (1) load paths are non-intuitive due to complex design domains or boundary conditions, or (2) the design process is to be automated to minimize effort associated with experience-based design. However, practically relevant boundary conditions are often difficult to put into a formal mathematical language to, for example, either feed it into a topology optimization algorithm, or provide precise quantitative criteria for CAE-supported manual design. This paper presents a survey of three industry use cases and identifies three types of requirements: the first can be directly cast into parts of an optimization problem statement (∼ 40%), the second is considered indirectly by adapting the optimization problem without explicit reference to the requirement (∼ 20%), and the third is only assessed after the design is finalized (∼ 40%). For categories 2 and 3 we propose directions of improvement to support formulating complex design tasks as unambiguous design problems.
Characterising a socio-technical system by its underlying structure is often achieved by cluster analyses and bears potentials for engineering design management. Yet, highly connected systems lack clarity when systematically searching for structures. At two stages in a clustering procedure (pre-processing and post-processing) modelled and external information were used to reduce ambiguity and uncertainty of clustering results. A holistic decision making on 1) which information, 2) when, and 3) how to use is discussed and considered inevitable to reliably cluster highly connected systems.
Subjective product quality is a fundamental aspect to maintain a high level of user acceptance and to provide a good user experience. Prototypes can be used to evaluate subjective product quality in early design phases. We conducted an empirical study to examine the influence of different (re)presentations of prototypes determined by the chosen technology (2D image, VR, AR, 3D print) and the user's technology acceptance. Based on the results we recommend 2D images as most reliable for evaluating subjective product quality.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.