As the industrial sector enters a phase of profound transformation, Generative AI is emerging as a strategic technological capability, enabling manufacturing enterprises to fundamentally reshape their operating models. From optimizing product design and enhancing factory performance to building agile and adaptive supply chains, Generative AI plays a pivotal role in strengthening competitiveness and driving sustainable innovation.
According to Amazon Web Services (AWS), implementing Generative AI is not merely about adopting a new technology—it requires a systematic approach built on a robust architectural foundation. This includes ensuring data security, optimizing processing performance, maintaining operational stability, and effectively controlling costs—critical factors for delivering real value in large-scale and complex manufacturing environments.
In a modern Digital Thread architecture, data from systems such as CAD, PLM, MES, ERP, and IoT are seamlessly connected, forming a unified data backbone across the entire product lifecycle. On top of this foundation, Generative AI acts as an “intelligence layer,” transforming raw data into actionable insights and supporting decision-making. When integrated with PLM systems like PTC Windchill, AI can deeply understand the engineering context of a product—from BOM structures, configurations, and versions to the full history of changes.
On the infrastructure side, Amazon Web Services provides services such as Amazon S3 (data storage), AWS Glue (data integration), and Amazon Redshift (data warehousing) to centralize and synchronize data from multiple sources. Building on this data foundation, Generative AI—through platforms like Amazon Bedrock or Amazon SageMaker—can analyze and generate real-time insights and predictions. As a result, enterprises move beyond merely “seeing data” to actively leveraging it for engineering operations and production decision-making.
In modern design environments, the combination of PTC Creo and Generative AI enables engineers to automate the generation and optimization of design alternatives. However, the core value lies not only in geometry creation, but in effectively leveraging accumulated engineering data—including design history, technical constraints, and operational know-how—centrally managed within PLM systems. This ensures that AI-generated design options are not only technically optimized but also aligned with real-world manufacturing conditions.
On the infrastructure side, Amazon Web Services significantly enhances computational capabilities, enabling efficient processing of complex simulation and optimization workloads. Resource-intensive tasks such as CAE, CFD, and design optimization can be deployed on Amazon EC2 or high-performance computing (HPC) clusters, substantially reducing processing time. At the same time, all design data is stored and version-controlled within PTC Windchill, ensuring that every AI-generated output adheres to configuration control and is ready for integration into downstream processes such as manufacturing and supply chain management.
One of the key models in modern manufacturing is “closed-loop engineering,” where operational data is fed back into the design phase to enable continuous product improvement. This represents a shift from fragmented approaches to a tightly integrated system that connects real-world production with engineering activities.
On the infrastructure side, Amazon Web Services provides services such as AWS IoT Core and AWS IoT SiteWise, enabling real-time data collection directly from machines and production lines. This data is processed through streaming and event-driven services like Amazon Kinesis or AWS Lambda, and then centrally stored in a data lake for further analysis.
When integrated with PLM systems such as PTC Windchill, Generative AI can analyze operational data, correlate it with product configurations, and trace the root causes of issues. For example, an abnormal vibration detected during operation can be linked to a specific design change made in PTC Creo.
Beyond issue detection, AI can also recommend design improvements or adjustments to manufacturing processes. This enables enterprises to move from reactive maintenance to a more proactive model driven by predictive insights and engineering recommendations—enhancing system reliability and optimizing operational performance.
(To be continued)
