The Shift Toward Autonomous Production in Process Industries | SPONSORED
The definition of autonomy in manufacturing is shifting from a conceptual goal to a practical requirement. According to Axel Lorenz, CEO of Process Automation at Siemens AG, autonomous production allows facilities to maintain consistent output despite volatile inputs. This capability is becoming critical as manufacturers face fluctuating feedstock quality and variable energy supplies from renewable sources.
The transition toward autonomy is not about replacing human oversight entirely. Instead, it is about enhancing Advanced Process Control (APC) with software-defined automation and Large Language Models (LLMs) to create flexible, self-optimizing systems.
Three Pillars of Autonomous Infrastructure
To move beyond traditional automated loops toward true autonomy, manufacturers must integrate three core technological enablers.
- Software-Defined Automation: Moving away from rigid hardware constraints allows for the flexibility needed in autonomous cells. This approach enables modules to adapt to changes in inbound logistics and process requirements in real time.
- Comprehensive Digital Twins: A digital twin must be more than a 3D visual. It must encompass the physics of the plant, the characteristics of the media flowing through it, and the specifics of the end product. An “evergreen” twin that learns from live data is essential for remote or unmanned operations.
- Data Accessibility via AI: LLMs are being used to synthesize vast amounts of operational data, making it accessible to both the control systems and the human operators who manage them.
Real-World Applications: From Offshore Safety to Polymer Efficiency
The business case for autonomy is already visible in high-stakes environments. On offshore platforms in the North Sea, the implementation of self-learning digital twins has enabled more onshore control of offshore assets. This shift reduces the need for helicopter transport and personnel on-site, directly lowering operational costs while increasing safety.
In terrestrial manufacturing, the benefits are equally tangible. For example, LyondellBasell utilized AI and LLMs to optimize polymer production. While traditional APC handles stable production well, changing polymer grades usually introduces significant downtime or instability. By integrating AI with existing control systems in a brownfield environment, the company achieved grade transitions 30% faster. This project delivered measurable value in less than a year without requiring a full plant overhaul.
The Path for Small and Mid-Sized Manufacturers
Autonomy does not require an “all or nothing” investment. Small companies and startups are using these tools to compete with established players by focusing on specific high-impact areas.
Modular Engineering
Hydrogen startups use digital twins to reduce costs during the tendering and engineering phases of modular plants. By reusing digital assets from the design phase through to operation, they shorten the time to market.
Operator Support
The safest first step for many manufacturers is “operator augmentation.” Using tools that allow operators to query the control system in natural language helps them understand the facts behind specific machine behaviors. This reduces the learning curve for new staff and minimizes human error in complex processes.
Future Outlook: Generative AI and Integrated Control Systems
The next phase of industrial productivity will be defined by the integration of AI into the core architecture of process control. Systems like Siemens PCS neo are already incorporating these technologies to bridge the gap between the installed base of legacy hardware and the future of autonomous software.
As feedstock variability and energy transitions continue to challenge the process industry, the ability to automate decision-making at the edge will become the primary driver of competitive advantage.
This article was written based on the video interview with Axel Lorenz, CEO of Process Automation at Siemens AG, recorded at the ARC Executive Forum 2026 in Orlando.
About the author
Lucian Fogoros is the Co-founder of IIoT World.
Sponsored by Siemens