Dez 09 2024

Unlocking Sustainable Manufacturing and Supply Chains: The Power of Semantic Technologies

by Ralf Hartmann on behalf of the AIMS5.0 consortium

The AIMS5.0 EU funded project stands for “Artificial Intelligence in Manufacturing leading to Sustainability and Industry5.0” and is a Collaborative Innovation Action between multiple domains with a consortium of over 50 academic and industry partners. 

Its main goals consist of strengthening the European digital sovereignty in comprehensively sustainable production, and boosting the economy by adopting, extending and implementing AI-enabled hardware and software components and systems across the whole industrial value chain. Its aim is to create human-centric workplace-conditions and a climate-friendly production. The AIMS5.0 project supports AI-enabled fabs to be more productive and eco-efficient. It advocates shorter supply chains, better resilience, global competitiveness and higher sustainability. 

The global semiconductor manufacturer Infineon Technologies AG has several contributions to the AIMS5.0 funded project, with one of them being Work Package Four. Work Package Four, “The Open Access Platform” is co-lead by Infineon Technologies and FernUniversität Hagen with contributions from DAC.digital and NXP. The primary objective of the Open Access Platform is to create a consistent and modular semantic description of the semiconductor industry containing industrial domains and supply chains based on the Digital Reference Ontology (DRO), which will be elaborated in the following.

 

The Semantic Web and Ontologies

To define the concept and strengths of ontologies, one first needs to understand the Semantic Web. The Semantic Web was coined from T. Berner Lee as an extension of the World Wide Web that enables machines to understand and interpret meaning behind data. It is founded on the concept of data interoperability allowing information to be easily shared, reused, and combined across different organizations and systems. As a result, a more connected and integrated digital ecosystem is enabled. Traditionally, data on the web has been structured in a human-centric manner, with information presented in formats like HTML text documents that are easily comprehensible to humans. However, these formats are not machine-friendly, making it challenging for computers to interpret and process the data efficiently. 

In contrast, the Semantic Web provides a new layer to the traditional web to explicitly define and structure information, allowing machines to automatically discover and integrate data from diverse sources and infer relationships by making logical connections. Therefore, human consumable data becomes machine consumable content, effectively bridging the gap between human understanding and machine processing, and unlocking new possibilities for automation and insight. For Search Engines, Selling Platforms or even for LLMs (Large Language Models) in a B2C environment, Semantic Web technologies are widely used. However, due to the strong need of high-quality processes, it is not common in the B2B market, even though the Semantic Web is crucial for seamless data exchange. With the help of well defined and verified ontologies, the required quality can be brought to certain domains in B2B – in our case for the domain of semiconductor and supply chains containing semiconductors. 

In essence, an ontology is a formal and explicit representation of knowledge that defines a set of concepts and relations as well as rules of a particular domain, enabling machines to understand and process the knowledge. By providing a common understanding of concepts and terms, data can be better integrated, searched for, and retrieved, since ontologies facilitate the understanding of relationships between concepts. This leads to reduced ambiguity, and therefore to more efficient decision making and problem solving. 

Utilizing all of those benefits, the Digital Reference Ontology (DRO) for the domain semiconductor and supply chains containing semiconductors plays a crucial role in exploring the opportunities of the Semantic Web. The DRO is an effort that has evolved throughout several funded projects with its initial start in 2019 with the Project Productive4.0 with 250 classes. Today, containing almost 2000 classes (status 2024), the DRO is actively used in AIMS5.0 and SC4EU. The classes are distributed over 14 lobes in one large ontology model – the DRO for the domain semiconductor and supply chains containing semiconductors – which can be seen in the picture headlining this article.   

 

DRO used in AIMS5.0 and the Open Access Platform

With the introduced concepts of the Semantic Web and the DRO as background knowledge, we can include the novelty of AIMS5.0. The work that collaborators are currently working on in AIMS5.0 Work Package Four is the Open Access Platform (OAP), a software application that will enable the project consortium to interact with the Digital Reference Model with their own specific use case, not only providing more potential classes to the DRO but also enabling use case queries to be answered. 

The OAP is designed to streamline data exchange within industrial supply chains, with a primary emphasis on the semiconductor and connected supply chain sectors, as well as facilitating seamless data sharing between these industrial domains and relevant markets. The ultimate goals of the OAP include achieving a significant reduction in carbon emissions, mitigation of the global semiconductor shortage, and minimizing the bullwhip effect.

Work Package Four aims to not only target Semantic experts to use the DRO, but also partners who have little to no semantic knowledge through the integration of existing and tested tools into one platform. The tools support the creation of ontologies from conceptual diagrams without extensive knowledge on the process, and the visualization of the results to directly present the created ontology. Through usage of the tools, the OAP will enable the user to generate ontologies which then can be connected to the correct lobe and offer solutions for additions to the DRO.