Integrating the Mechanical Domain into Seed Approach

Abstract

Data-driven technologies have found their way into all areas of engineering. In product development they can accelerate the customization to individualized requirements. Therefore, they need a database that exceeds common product data management systems. The creation of this database proves to be challenging because in addition to explicit standards and regulations the product design contains implicit knowledge of product developers. Hence, this paper presents an approach for the semantic integration of the engineering design (SeED). The goal is an automated design of an ontology, which represents the product design in detail.SeED fulfils two tasks. First, the ontology provides a machine-processable representation of the products design, which enables all kind of data-driven technologies. Among other representations, the ontology contains formal logics and semantics. Accordingly, it is a more comprehensible solution for product developers and knowledge engineers. Second, the detailed representation enables discovering of intrinsic knowledge, e.g. design patterns in product generations. Consequently, SeED is a novel approach for efficient semantic integration of the product design.

Model and Knowledge Representation for the Reuse of Design Process Knowledge Supporting Design Automation in Mass Customization

“…Design Parameters are used as targets, while other concepts of SeEDMC are stated in the annotations' bodies. As this meets requirement four, the fifth and last requirement is achieved by automatically transferring the design parameters and the annotations into the SeEDMC using a mapping scheme and design wrapper [14]. The feasibility has been probed through the implementation of a mass customization process for an individual bike crank.…”

“…In most development environments, the stored information can be extracted via an API. Particularly, in the mechanical domain, the SeED approach is used [14]. As stated previously, states define snapshots of the product at certain points in time.…”

“…SeEDMC expands the SeED approach. It processes geometric information only using the design process and corresponding concepts and axioms [14]. Within this important extension, semantic descriptions are used to integrate different aspects (e.g., geometry, functions, principles) of engineering design semantically, which is vital for design automation in mass customization.…”

“…Figure 6.Illustration of the extraction of structural data (e.g., shape feature) based on[14] and expansion for the extraction of annotated semantic data by the use of the annotation scheme.…”

Integration of product development data for further ontological utilization

“…For the further use of volume models, the National Institute of Standards and Technology (NIST) provides a tool called OntoSTEP, available as a Protégé plugin, which semantically extends the data in the standard STEP exchange format and translates it into an OWL-DL (OWL description logics) based on the creation of a terminology (TBox) and an assertion component (ABox) (Barbau et al, 2012). The SeED approach similarly pursues the semantic integration of volume model data into an ontology, built up in different steps to decompose and segment the model at hand (Dworschak et al, 2019). In this approach, the data is represented in a 1D array and stored as an xlsx file, which can preferably be integrated into the ontology using the previously presented Cellfie plugin (Protégé Project, 2022).…”