Engineering organizations face increasing product complexity, growing software content, expanding regulatory requirements, and pressure to accelerate innovation without increasing overhead. Yet the knowledge required for sound decisions remains fragmented across requirements, systems, product lifecycle management (PLM), design environments, manufacturing systems, and historical program data. Leaders must improve speed, quality, and reuse while controlling the cost and risk of change.
The challenge is no longer creating more engineering data. It is transforming trusted product and lifecycle context into engineering intelligence that helps teams make better decisions, move faster, and manage complexity at scale.
From the digital thread to AI-powered engineering intelligence
Agents are only as effective as the engineering context they can access. For manufacturers, that context resides within the digital thread connecting requirements, product structures, design decisions, validation evidence, engineering changes, manufacturing information, and lifecycle decisions. When that thread is governed and accessible, agents help teams understand how information relates to the product and the decision at hand.
This is what distinguishes engineering agents from general productivity copilots or traditional workflow automation. Engineering agents must operate within structured product data, configuration rules, lifecycle states, requirements traceability, and governance controls. They can assemble context across these sources, identify dependencies, and support decisions while engineers remain accountable for the outcome.
Consider an engineer evaluating a design change. The decision may affect product structures, requirements, manufacturing constraints, prior designs, and downstream processes. An engineering agent can bring relevant evidence together and expose relationships that otherwise require manual investigation. Engineering leaders routinely balance cost, quality, performance, risk, and schedule. Agents can surface tradeoffs, enabling faster, better-informed decisions without removing expert judgment.
The opportunity is not simply task automation. Engineering intelligence can expand capacity by reducing time spent assembling context and allowing experts to focus on tradeoffs and innovation. It can preserve lessons from previous programs, make knowledge reusable across teams, and strengthen impact analysis as changes move through the lifecycle.
A governed foundation for engineering agents
Delivering this value requires more than connecting a LLM or SLM to engineering files. Agents need trusted product context, lifecycle governance, and a framework for operating within engineering processes.
Built on Microsoft Azure, Aras combines PLM with engineering AI through InnovatorEdge AI, a governed framework for embedding agents into engineering workflows. Manufacturers can manage agents across the digital thread, allowing them to work with structures, requirements, configurations, and engineering knowledge in context. This creates a foundation for engineering intelligence while extending investments.
The strongest value emerges in engineering decision support. Agents can help teams retrieve knowledge, understand product dependencies, evaluate options, and apply previous decisions to new programs. They can also support engineering change management by improving impact visibility, traceability, and lifecycle control, where incomplete context increases downstream cost and risk.
Turning variant knowledge into decision support
SICK AG, a provider of sensor-based solutions for factory, logistics, and process automation, sought to improve transparency across a product portfolio, identify reuse and harmonization, and make product knowledge easier for teams to apply across programs. Achieving those goals required making variant logic across product bills of material accessible and actionable.
SICK partnered with Aras and Microsoft to connect that knowledge through a governed digital thread. Aras InnovatorEdge AI, powered by Microsoft Foundry and Azure, brings agentic capabilities to trusted PLM data while preserving the context and controls required for industrial engineering.
By analyzing existing 100% BOMs, the Variant BOM Agent can surface product features, options, dependencies, and configuration rules, then help engineers reconstruct variant logic and derive a consolidated 150% BOM. This gives teams a clearer foundation for evaluating commonality, reuse, harmonization, and product decisions.
The agent makes embedded product knowledge more accessible rather than treating BOM analysis as an isolated automation task. Engineers can review relationships and recommendations within governed workflows, applying their expertise to the resulting decisions. The approach supports transparency and decision speed while maintaining lifecycle context, traceability, and human accountability.
Specific productivity and return-on-investment figures are still being validated. SICK demonstrates how trusted product context can enable engineering analyses difficult to perform at scale. Its contribution is engineering decision support, with relevance to change impact analysis and portfolio decisions. It does not need to represent every engineering scenario to show value.
Scaling engineering intelligence across the enterprise
Successful agentic engineering begins with trusted context, but the long-term opportunity extends beyond automating individual tasks. Manufacturers can create an environment where engineering knowledge becomes continuously available, reusable, and scalable across the enterprise.
In that environment, agents do not replace engineering expertise. They amplify it, helping organizations manage complexity, accelerate development, improve reuse, preserve knowledge, and reduce the cost and risk of change. By combining Microsoft’s agentic technology, Aras product intelligence, and SICK engineering expertise, the collaboration shows a path toward engineering intelligence, enabling faster innovation while maintaining the governance, traceability, and control for industrial product development.
Step into the Factory of the Future
A connected and intelligent factory can help you improve operational resilience.
