How AI can
transform PLM

AI provides solutions to common PLM challenges

For example, it can:

Significantly improve the development of digital products such as PLM systems.

Parse large volumes of PLM data to extract meaning and correlations, overcoming the complexity of data and relationships and keeping track of versions, configurations, and so on.

Promote PLM adoption and usage by efficiently exposing knowledge from tangled sources of information.

Create industrial assets to rapidly populate a smarter design space and efficiently produce simulation results.

The result will be a new kind of PLM: a decentralized, data-centric PLM that supports flexibility, autonomy, and cooperation. Thanks to business logic decentralization, agents – whether human or AI – will be able to maintain consistent relationships across systems and across the data lifecycle, with no need for data to be stored centrally.

The power of AI in PLM is categorized in four pillars:

Agentic AI in
collaborative PLM

Enterprise lifecycle
integration / SW & HW synchronization

BOM quality and
digital continuity

AI-driven MBSE
enhancement

1. Agentic AI in collaborative PLM

Agentic AI enables flexible, scalable stakeholder interactions within collaborative PLM foundations, improving teamwork and responsiveness. As Marvin Minsky noted, “The parts make the whole. Many mindless agents make the mind.”

Industries and organizations worldwide face challenges in developing complex and diverse products, assets, and systems.

Success depends on strong collaboration between numerous disciplines, supported by new digital solutions.

Also needed is a new style of interaction, with looser coupling between data and models enabling more agile interdisciplinary collaboration services.

Here, agents have an important role to play.

2. Enterprise lifecycle integration /

SW & HW synchronization

AI can enable tight coordination between lifecycle phases and activities, which is critical for achieving overall product optimization, rather than just optimizing individual phases. Coordinating hardware and software activities by integrating software lifecycle management and PLM is one example; another is product/manufacturing coordination.

Several of our clients are investigating the possible use of AI in operations to facilitate feedback of service data (e.g. service BOMs and maintenance tickets) to engineering. This data could drive improvements to existing products and inform the design of future ones.

3. BOM quality and digital continuity

AI can improve the extended BOM to enhance quality and traceability, as it is able to parse large volumes of data and to identify correlations, causations, and relationships that might be missed by the human eye or by older automated methods. Machine learning (ML) can extract hidden patterns from product stores, and generative AI can expose the results in a meaningful and actionable form. AI agents can then further accelerate the verification & validation (V&V) of assets.

This reduces the errors that can occur downstream with a multi-view approach (multiple BOMs, each with its own list of required resources). AI can eliminate the risk that, for example, a part that is in the EVIEW (EBOM + metadata) gets omitted from the MVIEW (MBOM), resulting in a missing physical part.

We’re working with a client in the aerospace sector to apply AI to the task of ensuring that V&V plans are comprehensive, as well as using it to generate test cases. AI is also helping to verify that requirements are met across all BOMs.

From document-centric to data-centric.
From data-centric to agentic.

4. AI-driven MBSE enhancement

AI augments model-based systems engineering (MBSE) by automating tasks, enhancing decision-making, and improving traceability in engineering PLM. AI-powered MBSE enables organizations to evaluate architectural alternatives earlier, anticipate social and market trends, and optimize CapEx investment choices with significantly more confidence.

With clients in industries that deal with many thousands of product variants, such as automotive, we're working to show how AI can identify which variants are of commercial interest, so that they can focus on what matters most to the business.