From connected data to industrial intelligence

By: Kathryn Gerardino-Elagio

Artificial intelligence was a major theme at Siemens Realize LIVE Asia-Pacific 2026, but the discussion in Bengaluru went beyond adding copilots to existing software. Across the general sessions, technology demonstrations and media briefing, Siemens executives described a broader transformation in which industrial software is becoming more connected, contextual and capable of supporting decisions throughout the product lifecycle.

Held on 4–5 August, the event brought together more than 500 Siemens customers, as well as engineers, manufacturing executives and technology partners. More than 90 sessions covered design and engineering, simulation, product lifecycle management (PLM), manufacturing operations, quality and service.

The central message was that industrial AI must be grounded in reliable engineering and operational data. A general-purpose AI model may produce a convincing answer, but manufacturers need to know where the information came from, how it relates to their products and processes, and whether the result can be trusted.

Turning complexity into an advantage

During the general session, Tony Hemmelgarn, President and CEO of Siemens Digital Industries Software, explained how AI, digital twins and connected data can help companies manage increasing industrial complexity.

Modern products combine mechanical systems, electronics and millions of lines of software. Manufacturers must also respond to changing customer requirements, supply-chain disruption, sustainability targets and pressure to shorten development cycles.

A comprehensive digital twin allows engineers to model a product, production system or factory before making a physical investment. They can compare alternatives, predict performance and identify problems before tooling is ordered or equipment is installed. AI can then help users interpret the large amount of information generated during these activities.

Industrial software is therefore moving beyond documenting decisions after they have been made. It is increasingly helping manufacturers make better decisions earlier, when changes are less costly and easier to implement.

This idea continued during the media briefing, which brought together Siemens executives responsible for PLM, simulation, AI, manufacturing operations, low-code development, sales and regional management.

PLM provides context for AI

Joe Bohman, Executive Vice President of PLM Products, is responsible for software covering product design, engineering and manufacturing. From this perspective, PLM is becoming more than a system for managing drawings, bills of materials and product revisions.

PLM provides the structure that allows companies—and their AI systems—to understand relationships between requirements, components, manufacturing processes and suppliers. This context is especially important when AI is expected to work across departments rather than inside a single application.

Siemens describes several layers of industrial AI. Engineering copilots assist users within design and simulation tools. An AI fabric connects information from systems such as PLM and enterprise resource planning, while digital-thread agents can coordinate tasks across applications.

These capabilities could reduce the time engineers spend searching for information, transferring data and completing repetitive work. They could also help companies retain valuable knowledge when experienced employees retire or change roles.

However, AI cannot provide a complete picture if engineering, manufacturing and business information remains divided among disconnected systems. A strong digital thread is therefore essential.

Combining AI with engineering knowledge

Sam Mahalingam, Executive Vice President of Simulation, High-Performance Computing and AI, leads technologies that form an important part of Siemens’ industrial AI strategy.

Simulation already enables engineers to test how products and processes will perform without building every physical variation. When combined with AI, it can shorten calculation times and allow teams to evaluate many more design alternatives.

Physics-informed AI is particularly promising because it combines machine learning with engineering models and physical principles. Instead of relying only on patterns found in historical data, the system works within known technical constraints. This can improve confidence in applications where performance, safety and accuracy are critical.

Siemens is also bringing together data, models, workflows and domain knowledge through Intelligence Center X. The aim is to turn scattered industrial information into governed intelligence that can be used across engineering and manufacturing.

This addresses a common weakness in enterprise AI projects. Even a powerful model cannot create value by itself. It needs access to trusted information, connections to industrial software and an understanding of the process in which its recommendations will be used.

Connecting engineering with the factory

The event also made clear that the digital thread must extend beyond product development and into production.

Volker Albrecht, Head of Portfolio Management for Operations Software and CEO of Siemens Digital Logistics, represents the part of the portfolio that connects engineering plans with manufacturing, quality and logistics.

Manufacturing operations management systems link enterprise planning with activity on the shop floor. They coordinate production schedules, materials, work instructions, quality records and equipment data. When connected with PLM and automation, they help ensure that factories produce the correct product version using the approved process.

Siemens’ operations software strategy brings engineering, planning, execution and automation into a more unified environment. Opcenter X provides a cloud-oriented path that allows manufacturers to introduce new capabilities gradually while continuing to use established systems.

This is important for existing factories, where replacing every system and production asset at once is rarely practical. Manufacturers need to modernise in stages without interrupting operations.

AI can support this process by improving production scheduling, detecting quality issues and helping workers investigate manufacturing problems. In logistics, it can assist with forecasting, inventory, routing and the earlier identification of supply risks.

Filling workflow gaps with low code

Koert Hagoort, Senior Vice President of Product and Technology for Mendix, brings low-code application development into this connected environment.

Manufacturers often have gaps between their major enterprise systems. Employees may still rely on spreadsheets, emails or paper documents for inspections, approvals, maintenance requests and production changes.

Low-code tools allow companies to build applications around these processes without developing each solution from the ground up. Generative AI can further assist users in creating interfaces, defining workflows and connecting information.

This gives manufacturers greater flexibility to address the needs of a particular factory or production line while maintaining oversight from enterprise IT.

Turning technology into business results

Technology alone does not guarantee transformation. Robert “Bob” Jones, Chief Revenue Officer and Executive Vice President of Global Sales and Customer Success, has highlighted the gap between the large amount of industrial data collected and the relatively small proportion companies use.

Closing that gap requires manufacturers to connect information, update established workflows and help employees understand how new tools support their work.

Mathew Thomas, Managing Director of Siemens Digital Industries Software in India, also placed these developments within India’s expanding role in global industry. With its large technical workforce and growing automotive, aerospace, electronics and industrial machinery sectors, India is becoming an important centre for product development, software and manufacturing innovation.

Realize LIVE Asia-Pacific 2026 ultimately presented a future in which design, simulation, PLM, manufacturing operations and automation no longer function as separate layers. AI will operate across this environment, helping people find knowledge, evaluate alternatives and coordinate increasingly complex work.

For manufacturers, the starting point does not have to be a large AI programme. It can be a costly production problem, a slow approval process or an engineering task that consumes too much specialist time. The key is to connect the relevant information and apply AI where its impact can be measured.

The software may be becoming more intelligent, but the real advantage will come from how effectively manufacturers connect it with their people, processes and production systems.