Faster, Smarter, Trusted: How three pillars help put industrial AI to work
Industrial artificial intelligence (AI) is moving beyond proof-of-concept implementations and pilot projects for many industrial companies. As companies move to the next stage, the focus will evolve from the capabilities of individual AI models to systematic deployment across engineering tools, operations software and more. For Siemens, that means building industrial AI around a clear, three-part framework: faster engines, smarter execution, and trusted outcomes.
In a recent episode of the Future Ready Podcast, Samir Desai, Senior Director for AI and Data Strategy at Siemens Digital Industries Software, joined host Spencer Acain to walk through a three-part framework guiding the construction of industrial AI at Siemens. Samir and Spencer discuss how the framework is being put into practice with concrete examples and a candid look at what it takes to deploy AI responsibly in industrial environments.
Listen to the full episode below or continue reading for a summary of the discussion.
Faster Engines Accelerate Tools
The first pillar focuses on raw speed. Industrial AI can increase the computational efficiency of various software tools, often by orders of magnitude. Samir offered two examples to illustrate the potential gains.
First, ARM, a Siemens EDA customer, designs chips to meet Six Sigma quality standards. Ensuring a failure rate of no more than one in a billion has traditionally required a months-long process of running billions of Monte Carlo simulations before a design could move to production. Now, AI-powered predictive models embedded in IC verification tools have reduced that number to around 2,000 simulations, taking only hours to complete.
The second example, from Siemens Energy, focuses on the production of gas turbine blades that operate in extreme conditions that can exceed 3,500˚F, greater than the melting point of most metals. Extensive simulations are required to ensure precise thermal management and strength even under intense heat and pressure. Using AI-enhanced simulations, Siemens Energy has saved over 15,000 hours of computational time and achieved cost savings of $1.6 million per engine manufactured. Across a fleet of hundreds of engines, those savings compound significantly.
Smarter Execution through Industrial AI Assistance
The second pillar addresses the speed and efficiency of the engineers themselves. Samir described three distinct mechanisms through which Siemens is delivering industrial AI in support of engineering workflows: copilots, agents, and in-app AI.
Copilots bring natural language interaction directly into the tools engineers already use. Rather than switching contexts or searching through documentation, an engineer working in NX, Teamcenter, or Opcenter can ask questions and receive guidance in plain language, in the moment, without leaving their workflow.
Agents take this a step further by performing actions autonomously on behalf of the user. Samir distinguishes between two types. Domain agents operate within a specific tool, executing tasks at the direction of the user and always keeping the human in the loop.
The second type, digital thread agents, operate at a higher level, orchestrating actions across multiple tools and systems. Using the example of a design change to an EV battery, Samir explained how a digital thread agent can coordinate the downstream effects of the change across the 3D design tool, the electrical design tool and the manufacturing operations platform.
At each stage the digital thread agent ensures the domain agent performs its part of the workflow in the right sequence, without requiring the engineer to manage the handoffs. Finally, in-app AI describes capabilities that are embedded directly into the software and operate in the background, without requiring any explicit user interaction.
Connected Data Produces Trusted Outcomes
The third pillar addresses the challenge of data fragmentation, possibly the most complex structural impediment to industrial AI. Engineers and decision-makers routinely work across multiple systems or platforms, including PLM, MES, ERP, CRM and more. Each holds data relevant to a common product or program, resulting in siloes that inhibit cross-system insight or analysis.
Samir illustrated the problem by returning to the EV battery example. If a design engineer needs to change a component, understanding the full impact of that change currently requires pulling subject matter experts from each impacted system into a meeting and working through the implications manually. It is slow, labor-intensive, and dependent on the availability of the right people at the right time.
The conversation turned to Siemens’ AI Fabric and how it helps overcome data siloes, supported by RapidMiner Graph Studio. Knowledge graphs create a contextualized fabric of data systems, representing where data is stored and interrelationships between these locations. With this fabric established, agents and copilots can be layered on top to enable engineers to ask cross-system questions directly. The EV battery engineer, for example, could check available supplier data to understand how a proposed design change may ripple through the rest of the vehicle system.
Building Trusted and Reliable AI for Industrial Users
Faster engines, smarter execution and trusted outcomes reflect a commitment to the development of robust, reliable and accurate industrial AI solutions. The goal is not to deploy AI for its own sake, without considering the foundation on which it is built. Instead, Samir describes a framework that will help companies move faster, develop better products and lower costs.
Check out the Future Ready Podcast to learn more about industrial AI, the Digital Twin, connected automation, and other key technologies transforming the future of industry.
Siemens Digital Industries Software helps organizations of all sizes digitally transform using software, hardware and services from the Siemens Xcelerator business platform. Siemens’ software and the comprehensive digital twin enable companies to optimize their design, engineering and manufacturing processes to turn today’s ideas into the sustainable products of the future. From chips to entire systems, from product to process, across all industries. Siemens Digital Industries Software – Accelerating transformation.
