How AI is making digital twins more accessible for Asia’s manufacturers

Alex Teo, Vice President & Managing Director, Southeast Asia, Siemens Digital Industries Software

Maintaining manufacturing competitiveness is becoming increasingly challenging as organisations navigate rising production costs and growing operational complexity amid disruptions to global energy and supply chains.

Recognising these pressures, governments across Asia have introduced strategies to accelerate industrial transformation. Malaysia’s New Incentive Framework 2026 rewards manufacturers that create high-value jobs, strengthen local supply chains, transfer technology, and invest in productivity, sustainability and innovation. Similarly, Singapore’s Manufacturing 2030 vision aims to strengthen the country’s position as a global advanced manufacturing hub by increasing manufacturing value-added by 50-percent.

Achieving these ambitions will require manufacturers to embrace digital technologies that improve productivity, resilience and sustainability. Yet many manufacturers still rely on conservative engineering approaches, effectively building in additional safety margins through larger equipment or more complex production systems instead of validating performance through simulation. While this approach may reduce perceived risk, it can also increase material consumption, engineering effort and operating costs.

Digital twins offer a more efficient alternative. By creating a virtual representation of products, production lines or entire factories, manufacturers can evaluate performance, identify issues earlier and optimise operations before making physical investments. The challenge, however, is that simulation has traditionally required specialised expertise, limiting broader adoption across engineering teams.

Making simulation more accessible

Compared with sectors such as automotive or aerospace, simulation remains underused across much of the industrial machinery sector. Many organisations still see it as something that requires highly specialised skills, or question whether the return justifies the investment.

In practice, the digital twin enables manufacturers to evaluate design decisions virtually before committing resources on the factory floor. Engineers can test multiple production scenarios, validate machine behaviour and optimise workflows in a virtual environment without disrupting live operations. Tasks that once required multiple rounds of testing can now be completed much faster, allowing organisations to adapt to changes more quickly when accelerating deployment.

Artificial intelligence is now making these capabilities significantly more accessible. AI helps lower the barriers to simulation by automating time-consuming tasks such as recognising boundary conditions, recommending material properties and configuring simulation models. This allows less experienced users to build and run simulations with greater confidence, while enabling experienced engineers to complete complex analyses more efficiently.

This becomes more valuable as manufacturers move towards executable digital twins (xDTs) that operate alongside physical assets in real time, this democratisation of simulation becomes even more important. By continuously comparing operational and virtual performance, xDTs can help detect anomalies earlier, optimise maintenance schedules and improve equipment reliability throughout the production lifecycle.

Unlocking greater value with AI

The value of simulation increases further when AI is integrated throughout the digital twin lifecycle.

One challenge with predictive maintenance is that machine failures occur relatively infrequently. While this is positive from an operational perspective, it also means manufacturers often lack enough failure data to train predictive models effectively. AI helps address this gap by generating realistic synthetic data that allows digital twins to simulate a broader range of operating conditions and potential failure scenarios.

This enables manufacturers to make better informed operational decisions, such as whether equipment can continue operating safely until replacement parts arrive or whether immediate intervention is required. Over time, each operational cycle feeds more data back into the digital twin, creating a continuous loop that strengthens engineering decisions, maintenance planning and future system performance.

Siemens’ recent Digital Twin Composer illustrates this shift. It combines high-fidelity digital twin data with real-world operational information in a secure virtual environment, enabling manufacturers to visualise, test and optimise products, processes and factories before physical implementation. PepsiCo, for example, used the technology to validate new factory configurations and improve throughput before making physical changes, demonstrating how virtual environments can support better operational decisions and reduce implementation risk.

Strengthening Asia’s manufacturing competitiveness

As Asia strengthens its position as a manufacturing hub, digital capabilities will increasingly determine how quickly organisations can respond to changing market demands. From electronics and semiconductors to automotive and precision engineering, manufacturers are under pressure to increase flexibility while maintaining productivity and controlling costs.

AI and digital twins provide the foundation for achieving this balance. By combining physics-based simulation with operational data, manufacturers can evaluate complex production scenarios, optimise machine performance and refine operations within a virtual environment before implementing changes on the factory floor.

The significance of AI is not simply that it makes digital twins more powerful. It makes them usable by more people. As simulation becomes easier to configure and apply, manufacturers can extend virtual testing beyond specialist teams and embed it more deeply across engineering, production, and maintenance. That broader access will be critical to strengthening Asia’s manufacturing competitiveness.