Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Industrial Semantics-Aware Digital Twins: A Hybrid Graph Matching Approach for Asset Administration Shells

Ariana Metović, Nicolai Maisch, Samed Ajdinović, Armin Lechler, Andreas Wortmann, Oliver Riedel

TL;DR

The paper tackles the problem of semantic misalignment across Asset Administration Shell (AAS) models due to heterogeneous vocabularies. It proposes a hybrid graph-matching pipeline that first applies SPARQL-based pre-filtering to prune structurally incompatible AAS and then uses RDF2vec embeddings to compute semantic similarity in vector space. The approach integrates symbolic reasoning with data-driven representations to enable content-level comparability and reuse of AAS submodels and metadata, addressing interoperability challenges in Digital Twin networks. This work aims to reduce manual engineering effort by improving asset discovery, reuse, and automated configuration across manufacturers and domains.

Abstract

Although the Asset Administration Shell (AAS) standard provides a structured and machine-readable representation of industrial assets, their semantic comparability remains a major challenge, particularly when different vocabularies and modeling practices are used. Engineering would benefit from retrieving existing AAS models that are similar to the target in order to reuse submodels, parameters, and metadata. In practice, however, heterogeneous vocabularies and divergent modeling conventions hinder automated, content-level comparison across AAS. This paper proposes a hybrid graph matching approach to enable semantics-aware comparison of Digital Twin representations. The method combines rule-based pre-filtering using SPARQL with embedding-based similarity calculation leveraging RDF2vec to capture both structural and semantic relationships between AAS models. This contribution provides a foundation for enhanced discovery, reuse, and automated configuration in Digital Twin networks.

Industrial Semantics-Aware Digital Twins: A Hybrid Graph Matching Approach for Asset Administration Shells

TL;DR

The paper tackles the problem of semantic misalignment across Asset Administration Shell (AAS) models due to heterogeneous vocabularies. It proposes a hybrid graph-matching pipeline that first applies SPARQL-based pre-filtering to prune structurally incompatible AAS and then uses RDF2vec embeddings to compute semantic similarity in vector space. The approach integrates symbolic reasoning with data-driven representations to enable content-level comparability and reuse of AAS submodels and metadata, addressing interoperability challenges in Digital Twin networks. This work aims to reduce manual engineering effort by improving asset discovery, reuse, and automated configuration across manufacturers and domains.

Abstract

Although the Asset Administration Shell (AAS) standard provides a structured and machine-readable representation of industrial assets, their semantic comparability remains a major challenge, particularly when different vocabularies and modeling practices are used. Engineering would benefit from retrieving existing AAS models that are similar to the target in order to reuse submodels, parameters, and metadata. In practice, however, heterogeneous vocabularies and divergent modeling conventions hinder automated, content-level comparison across AAS. This paper proposes a hybrid graph matching approach to enable semantics-aware comparison of Digital Twin representations. The method combines rule-based pre-filtering using SPARQL with embedding-based similarity calculation leveraging RDF2vec to capture both structural and semantic relationships between AAS models. This contribution provides a foundation for enhanced discovery, reuse, and automated configuration in Digital Twin networks.
Paper Structure (23 sections, 3 equations, 3 figures)

This paper contains 23 sections, 3 equations, 3 figures.

Table of Contents

  1. Introduction
  2. Fundamentals
  3. Asset Administration Shell
  4. Knowledge Graphs and Resource Description Framework
  5. State of The Art
  6. RDF Technology for AAS
  7. RDF Stores and Querying
  8. Comparability of RDF Graphs and Data
  9. Vectorial Matching Approaches
  10. Research Gaps
  11. Architecture
  12. Data Storage and pre-discovery via SPARQL
  13. Vectorization via RDF2vec
  14. Challenges and Implementation
  15. RDF2vec method
  16. ...and 8 more sections

Figures (3)

  • Figure 1: Example of the basic Structure of an Asset Administration Shell based on IDTAWorkstreamSpecificationofAAS
  • Figure 2: Overall workflow of RDF2vecPaulheim.2023
  • Figure 3: Concept of the the proposed Hybrid Graph Matching Approach