arXiv:2404.06571v1 Announce Type: new
Abstract: Sourcing and identification of new manufacturing partners is crucial for manufacturing system integrators to enhance agility and reduce risk through supply chain diversification in the global economy. The advent of advanced large language models has captured significant interest, due to their ability to generate comprehensive and articulate responses across a wide range of knowledge domains. However, the system often falls short in accuracy and completeness when responding to domain-specific inquiries, particularly in areas like manufacturing service discovery. This research explores the potential of leveraging Knowledge Graphs in conjunction with ChatGPT to streamline the process for prospective clients in identifying small manufacturing enterprises. In this study, we propose a method that integrates bottom-up ontology with advanced machine learning models to develop a Manufacturing Service Knowledge Graph from an array of structured and unstructured data sources, including the digital footprints of small-scale manufacturers throughout North America. The Knowledge Graph and the learned graph embedding vectors are leveraged to tackle intricate queries within the digital supply chain network, responding with enhanced reliability and greater interpretability. The approach highlighted is scalable to millions of entities that can be distributed to form a global Manufacturing Service Knowledge Network Graph that can potentially interconnect multiple types of Knowledge Graphs that span industry sectors, geopolitical boundaries, and business domains. The dataset developed for this study, now publicly accessible, encompasses more than 13,000 manufacturers’ weblinks, manufacturing services, certifications, and location entity types.
Manufacturing system integrators face unique challenges in sourcing and identifying new manufacturing partners in today’s global economy. Supply chain diversification and agility are crucial for reducing risk and enhancing efficiency. The use of advanced language models, such as ChatGPT, has gained significant interest due to their ability to generate comprehensive responses across various knowledge domains. However, these models often struggle with accuracy and completeness when it comes to domain-specific inquiries, especially in areas like manufacturing service discovery.
This research proposes a novel approach to address this challenge by leveraging Knowledge Graphs in conjunction with ChatGPT. Knowledge Graphs provide a structured representation of information that can be easily queried and analyzed. By integrating bottom-up ontology and advanced machine learning models, a Manufacturing Service Knowledge Graph is developed using a combination of structured and unstructured data sources.
One of the key advantages of this approach is the ability to leverage the digital footprints of small-scale manufacturers throughout North America. By incorporating information from their weblinks, manufacturing services, certifications, and location entity types, a comprehensive dataset is created. This dataset, which is now publicly accessible, contains information on over 13,000 manufacturers.
The developed Knowledge Graph and the learned graph embedding vectors can then be used to address complex queries within the digital supply chain network. The integration of Knowledge Graphs with ChatGPT allows for enhanced reliability and greater interpretability in the responses provided. This multi-disciplinary approach combines concepts from ontology engineering, machine learning, and natural language processing to create a powerful tool for manufacturing system integrators.
Furthermore, the scalability of this approach is highlighted, as it can accommodate millions of entities and potentially form a global Manufacturing Service Knowledge Network Graph. Such a network could interconnect multiple types of Knowledge Graphs spanning industry sectors, geopolitical boundaries, and business domains. This interconnectedness would further streamline the process of identifying manufacturing partners and enable manufacturers to tap into a global network of potential clients and collaborators.
In summary, this research presents an innovative method to enhance the capabilities of ChatGPT in the context of manufacturing service discovery. By integrating Knowledge Graphs and leveraging the digital footprints of small-scale manufacturers, this approach offers improved accuracy, completeness, reliability, and interpretability. The multi-disciplinary nature of this research showcases the potential of combining concepts from ontology engineering, machine learning, and natural language processing to address complex challenges in the manufacturing industry.