Intelligence is getting democratised in industries: Pankaj Vyas, Siemens

In an insightful interview with Pankaj Vyas, CEO and MD, Siemens Technology and Services, provides a comprehensive look into the company's strategic evolution in India. The conversation explores how Siemens has transformed its Indian operations from a mere "extended workbench" to a hub of innovation, ownership, and acceleration. 

Pankaj delves into the core of their work, revealing how they are leveraging AI, digital twins, and a unique IT-OT integration strategy to drive significant advancements in the industrial, infrastructure, and mobility sectors. He also offers a glimpse into Siemens' future-focused initiatives, including its commitment to sustainability and its approach to nurturing a culture of innovation.

Siemens Technology is transitioning from simply executing global tasks to taking ownership through IP and innovation. What is driving this shift, and how are you exploring core AI innovation and intellectual property creation worldwide?

The foundation for this shift is the rich domain and tech expertise we have developed over the years. Over the last five-plus years, our learning has focused on the complete stack, meaning helping our people grow their technology, product, domain, and business know-how. This has all come together to drive the innovation we are seeing.

On top of that, we have established Communities of Experts who focus on emerging areas like data and AI, cloud, edge computing, UX, and cybersecurity. The goal is to bring together all our domain and tech knowledge to create and propose more innovative solutions. Another factor working in our favour is that we started investing in digitalisation much earlier, even before the launch of Siemens Xcelerator, our digital business platform. Because we invested in these areas ahead of time, we are better positioned to take on more innovative projects and drive proactive innovations.

To drive innovation, we also had to demystify what it is. When we hire young talent or even experienced professionals, they are often used to a certain way of doing things. It was important for us to demystify innovation and get people to ask, "Can I innovate?" The answer is that anyone can. For this, we built a platform called Ideascope, where anyone can submit an idea, select mentors, and work within an innovation sandbox. This has led to very impressive results in our innovation trajectory.

As mentioned, Siemens works in multiple industries and has been creating industrial co-pilots. What measurable impact are these AI-powered human-machine interfaces creating in terms of productivity, efficiency, or addressing the skill gap in the industry?

When we look at AI, including GenAI, our focus is on two areas: product and service enhancement with AI, and improving productivity with AI. These are the two primary areas where AI is used.

The solutions that Siemens AG provides involve significant effort in commissioning and engineering at the physical site. This is very different from a pure software solution. A lot of human intervention is required, for instance, when people install and program PLCs in a plant or deploy building management systems and sensors. It is a very high-touch deployment and commissioning process.

One of our focus areas is to use AI and GenAI to make these processes much more efficient, so that our commissioning engineers and customers' teams are more productive. The second area is improving human-machine interaction, such as with the industrial co-pilot you mentioned, which is still in development. It is difficult to put a number on the impact right now because we are just scratching the surface. However, the early results from our internal use of these solutions show that we are saving tens of thousands of hours for the people who are commissioning and engineering solutions. This clearly shows we are moving in the right direction. We are also seeing a lot of positive outcomes in the quality of products by using AI in the design and development stages.

Beyond voice commands, what other forms of human-machine interfaces is Siemens exploring to make complex industrial systems more accessible and user-friendly?

Voice commands are just one part of the interface, but what makes it powerful is what the machine can do for you after that. For example, in a manufacturing plant, an HMI (Human-Machine Interface) panel is typically managed by operators who carry manuals and require extensive training. Imagine if the HMI panel could learn what is happening in the plant on its own and guide the operator. That's the kind of power we are talking about. We are moving toward infusing more intelligence into the machines themselves to make the operator's life easier. This also allows us to bring in new people and make them employable much faster because they will not need as much training. This also helps reduce human errors, as the intelligent machine can guide the operator.

Today, AI is moving beyond experimentation to real-life implementation. When you are developing industrial AI solutions, how do you ensure they are "industrial-grade" and meet the strict requirements for performance and scalability in these complex sectors?

Industrial AI has nuances that are very different from non-industrial AI. The amount of data we deal with is huge. An average smart manufacturing plant can generate anywhere from 5 to 10 terabytes of data. It is a huge task to build models that can analyse this data quickly and derive useful intelligence. We also have to determine what data is useful and what is not.

Another nuance is that despite having so much data, we often do not have enough data on failures, because you do not want to have a lot of failures. In these cases, we rely on synthetic data creation and simulation.

To ensure our solutions are robust, we understand that industries like Digital Industries, Smart Infrastructure, and Mobility will only adopt new technologies when they have a sense of safety. We cannot just start using AI without knowing the implications. So, before deployment, we start with monitoring. We monitor the data and connectivity to ensure they are solid. Then, we pilot the solutions in smaller areas and conduct field trials. This is an agile way of doing things. Once we see the desired results, we move to wider deployment. Before all this, we used digital twins to simulate the entire behaviour of a manufacturing plant in a digital world. This allows us to test our AI models thoroughly in a digital environment before deploying them in the real world.

Siemens is positioning itself at the intersection of IT and OT (Information Technology and Operational Technology). How are you leveraging AI to bridge these two traditionally separate worlds for real-time automation and decision-making?

Our focus is on adding value in this whole IT-OT integration. The industries Siemens AG operates in are adopting new technologies to make sense of the data and to become more efficient, safe, and flexible.

The first step in IT-OT integration is a solid connection: Is your OT system connected? The second step is having the right strategy for what happens with the data at the edge or in the cloud, depending on latency requirements. The third step is what you do with all this data, which is where analytics and AI come in.

Siemens Xcelerator, our digital business platform, has three pillars: portfolio, ecosystem, and marketplace. Within the portfolio, we focus on making our products connected, open, and interoperable, with a uniform communication protocol and cybersecurity in place. We are involved in connecting these devices, securing them, and developing solutions for the middleware on the edge (industrial edge, grid edge, etc.) and in the cloud. This completes the entire IT-OT integration stack. All of our Communities of Experts and the domain knowledge acquired over decades come together to provide impactful solutions for this integration.

As you work in mobility and manufacturing, sustainability is a key focus. How are you using your AI and digital twin platforms to help customers reduce their carbon footprint and achieve sustainability goals?

Siemens AG has a framework called DEGREE that focuses on several areas: Decarbonization, Ethics, Governance, Resource efficiency and circularity, Equity and Employment. We are making a significant contribution to decarbonization and resource efficiency.

For decarbonization, we start by monitoring energy utilisation in industries. Our studies show that using digitalisation appropriately can lead to approximately a 40% reduction in energy usage, which in turn reduces CO₂ emissions.

Across our industries, we are driving sustainability. In buildings, our solutions focus on comfort, safety, and sustainability. For example, a hospitality chain can monitor the carbon footprint and energy usage of all its hotels and even individual floors to spot wastage. In mobility, we improve resource efficiency by enabling more trips on the same track without laying new ones. By using smart planning and signalling, we can increase the number of trips and coaches while maintaining safety, which encourages more people to use environmentally friendly public transport.

We are also working on circularity. Our solutions are designed to address product emissions at the design stage, as 80% of a product's emissions are determined then. We also look at a product's entire journey, from "cradle to death," to ensure that once a product is no longer useful, its components are reused or recycled instead of being buried in landfills.

How is Siemens partnering with Indian academia and startups to align with your AI innovation roadmap?

We work closely with academia. We have M.Tech programs with BITS Pilani, my alma mater, in software engineering and data and AI for our employees. We also actively ensure the curriculum meets future needs. We engage with IITs and IISc to help them incubate new ideas. Our strategic priority is "technology with purpose," so we look for solutions with a real-world impact. Additionally, we provide our technology and domain expertise to students in certain institutes to help them in their labs and with their projects.

In terms of startups, Siemens AG has been acquiring both large and medium-sized companies. From an Indian perspective, we work with bodies like NASSCOM to find out what is happening in areas like software-defined automation, industrial edge, industrial AI, and energy conservation solutions.

What do you believe will be the most significant and disruptive emerging technologies and trends in the industrial sector shortly?

 In the industrial sector, the IT-OT integration, with the significant use of AI and GenAI, will have a huge impact. So far, we have talked about the infusion of technology, but now we are seeing the infusion of intelligence. Previously, intelligence resided on the cloud, and then it moved to the edge. Now, intelligence is being integrated into the machines themselves. We are seeing large and small language models getting closer to the machines, democratizing intelligence in industries. This will make machines a lot more intelligent and help people interact with them more smoothly and effectively. I think this movement of intelligence closer to the machines will be the most disruptive part of the future.