MFGSCE4: Digital thread: contextualization and knowledge graph
In most manufacturing organizations, the product lifecycle data is typically stored in enterprise systems such as Product Lifecycle Management (PLM), Enterprise Resource Planning (ERP), Manufacturing Execution Systems (MES) and Customer Relationship Management (CRM). Due to the disconnected nature of enterprise infrastructure, the generated data is often fragmented and unused.
Managing and contextualizing data across the product lifecycle is challenging due to variations in data usage depending on the specific stakeholder interacting with it. Manufacturing organizations continuously aim to address the connected enterprise outcomes through various strategic initiatives. One notable initiative is the digital thread for connected enterprise.
Digital thread refers to the seamless flow of data across the entire product lifecycle, from requirements, design, production, and eventually to service and disposal. This data-driven approach enables manufacturers to contextualize information and gain insights for better decision-making.
A comprehensive digital thread should incorporate the requirements of various stakeholders, offering enhanced contextualization and knowledge integration capabilities, both structured and unstructured, to unlock the potential of manufacturing data.
Knowledge graphs provide a great way to represent and connect data from various sources, contextualizing and improving the understanding of complex relationships between entities. They provide a structured way to organize and link data entities, making it simple to find and traverse relationships between information from various sources.
Contextualization is the process of adding relevant context to data, which can help derive meaningful insights. By contextualizing data, manufacturers can better understand the context and conditions surrounding specific events or data points, which better informs their decision-making.
ADD FIGURE CAPTION HERE
A well-architected manufacturing digital thread with contextualization and knowledge graph capabilities should support the following characteristics:
-
Data integration and ingestion: The ability to ingest and integrate data from various sources, such as requirements, design, planning, supply chain, manufacturing execution data along with the unstructured data, is crucial for building a comprehensive digital thread.
-
Contextualization: Incorporating contextualization enables manufacturers to enrich their data with meaningful context and uncover hidden relationships and patterns.
-
Scalability and availability: The digital thread should be built on a scalable and highly available infrastructure, capable of handling variable data volumes and workloads without compromising performance or reliability.
By contextualizing in the manufacturing digital thread, manufacturers can better understand their data, gain deeper insights, and drive continuous improvement across the product lifecycle. This data-driven approach improves decision making, increases efficiency, enhances product quality, and helps you innovate within your organization.
Digital thread solution framework
Data from various product lifecycle processes forms the foundation of this digital thread solution framework. The subsequent layer encompasses core enterprise systems, including PLM, ERP, and MES, which manage specific aspects such as people, processes, engineering, and manufacturing data within the enterprise.
The next layer is the connected data, which involves both knowledge graph and unstructured data. Together, they uncover insights and provide a comprehensive understanding of the interconnected manufacturing enterprise data.
Finally, large language models are integrated with the knowledge graph and unstructured data, creating advanced queries and accessing natural language capabilities.
ADD FIGURE CAPTION HERE
The solution aims to accelerate innovation by seamlessly connecting data from various systems and generate insights through the manufacturing digital thread framework. This framework establishes an intelligent structure where data is interconnected, empowering manufacturing stakeholders in their decision-making processes.
Digital thread reference architecture
The manufacturing digital thread solution architecture is implemented through the strategic integration of Amazon Neptune graph database, Amazon OpenSearch Service, and Amazon Bedrock, a fully managed generative AI service. The components are further enhanced by various AWS services, creating a comprehensive solution for the digital thread.
ADD FIGURE CAPTION HERE
-
Identify key stakeholders in the manufacturing organization and understand the business needs.
-
Identify data sources to build a digital thread on AWS
-
Ingest data into AWS using AWS Data Migration Service
for database migrations and AWS DataSync for large dataset movement. -
Upload ingested data into Amazon S3
for secure storage for further processing and analysis. -
Use Amazon Neptune bulk loader capability to ingest the data from Amazon S3
to Amazon Neptune graph database. -
Create chunks from unstructured documents stored in Amazon S3
, embed using Amazon Titan Text Embeddings V2, and store them in Amazon OpenSearch Service vector database. -
Select foundation models in Amazon Bedrock
. -
Link Amazon Bedrock
, Amazon Neptune , and Amazon OpenSearch Service , and integrate with AWS Lambda and Langchain. The orchestrator coordinates the process of generating the opencypher query using Amazon Bedrock Foundation models, runs the query against the Amazon Neptune graph, converts the graph query results into natural language using Amazon Bedrock , extracts the unstructured information from Amazon OpenSearch Service , and returns the relevant message to the user. -
Create an application layer using AWS Fargate
for container orchestration, Amazon Elastic Container Registry for managing container images, Elastic Load Balancing for efficient traffic distribution, Amazon Route 53 for DNS, and Amazon Cognito for authentication. -
Use Amazon VPC
to operate the application in a secure and isolated network. AWS Identity and Access Management enhances access control, while AWS Certificate Manager manages certificates and AWS WAF provides web application security. Malicious activity is constantly monitored by Amazon GuardDuty . The data at rest is encrypted with AWS Key Management Service and can be integrated with other third-party KMS solutions. -
Use AWS CloudTrail
to enhance transparency by tracking activities, Amazon CloudWatch to monitor resources, and AWS CloudFormation for automated resource deployment of digital thread application.
