AI has found its way into factories, where it is helping manufacturers make sense of data, improve efficiency, and protect workers. But making this work requires more than installing sensors or adopting new software. It involves connecting different systems, analysing the information they generate and turning it into decisions that can improve day-to-day operations. To understand how this is playing out in India, I spoke with Praveen Arora, Vice President, loT, Tata Communications, about the role of AI, the Internet of Things (IoT), computer vision, and connected wearables in manufacturing. We also discussed why worker safety matters for business performance and how AI could shape the factories of the future.
Praveen Arora: AI is one of the key technologies transforming the industry. While we have been talking about it for some time, the real value is being created at the intersection of IoT and AI. That is where the power comes in. If you look at an enterprise environment, data comes in from tens of thousands of sensors. Traditionally, IoT has been responsible for collecting and bringing this data together. Once the data is available on a platform, AI provides the analytical layer on top of it. This brings together data from sensors, enterprise systems, and applications to generate what we call Tata Communications Factory 360.
With this, we get a 360-degree view of how plant operations are running. This includes identifying inefficiencies on a production line, monitoring safety and power consumption, checking compliance, comparing plant capacity with actual output, and identifying potential threats and bottlenecks.
When millions of data packets come in from multiple systems, no single system can simply take all that information and act on it. There are so many parameters and variables to consider. This is where AI comes in. It helps process that information and generate meaningful outcomes that support business decisions. That is where we are seeing the real value of AI in enterprise environments.
Praveen Arora: When we look at safety, we provide solutions in the form of a complete system. As I explained earlier, this brings together AI and IoT in an enterprise environment.
For safety, we provide the network, our own device ecosystem, devices developed in-house, a platform and applications to process the information, and managed services to run the solution and deliver benefits to customers.
From a connected-worker safety perspective, data comes from devices that provide location information and sensor data. For example, immobility sensors can detect when a person has remained stationary for a longer period. This could mean that the person has fallen or is in some other situation requiring assistance. The data is analysed, and help can be dispatched or action taken immediately. The system also provides the person’s exact location through the connected ecosystem of devices, network, and platform.
Another application involves marking certain areas as hazardous or safe to enter. If a worker’s movement pattern indicates that they are heading towards a hazardous zone, the system can issue an alert to the employer, the worker, and the command and control centre so that appropriate action can be taken.
There is also a feedback mechanism in which alerts can be delivered through signals such as a buzzer, vibration, SMS, or email. We integrate these capabilities with our broader portfolio of solutions, including Interaction Fabric, so that information can be brought together through unified communications.
Praveen Arora: One of the key things is to put people at the centre of operations. Whatever you want to automate or digitalise, people remain at the centre of it. Our belief is that if you take care of people, the rest of the operations can benefit as well. That is why we focus on offerings around safety and sustainability.
Take continuous operations in the manufacturing sector as an example. If an incident happens, the entire production process can stop. Even if production is halted for just 20 or 30 minutes, bringing the process back to the same level of efficiency can take hours. The entire chain in a process industry can be affected, and that results in a loss of production. So, when you focus on people, you are also focusing on operational efficiency, apart from taking care of the individual.
My message to manufacturers is that safety should be treated as paramount. There are different ways to address it using technology. This could involve physical safety measures, wearables, or video analytics.
AI and machine learning engines can be used for safety incident monitoring. They can help identify situations such as spills or fires, check whether workers are wearing the required safety equipment, and verify whether their harnesses are in place when they are working at height.
The complexity of these use cases can help prevent serious and even fatal incidents. I believe safety is something all manufacturers should take seriously.
Praveen Arora: Over the next four to five years, there is going to be a significant shift in how AI is used. If you look at the transition from Industry 3.0 to Industry 4.0 in India, many of these changes have already happened. A lot of data has been generated, and many systems have been incorporated into manufacturing environments. Now is the time to use that data more efficiently and effectively.
This could involve improving overall productivity or increasing capacity. Instead of being limited to the existing capacity, you should be able to improve it further. That is the power of bringing everything together.
For example, in one plant, there were around 13,000 sensors, 20 different systems, and six to seven different types of controllers, along with a manufacturing execution system. Data was coming in as millions of packets spread across different servers.
Each of these systems was generating its own specific output. But when the data was integrated, it became possible to get a broader picture of how the plant was operating. This helps manufacturers understand and manage capacity and efficiency through a single, integrated view of their operations.
That is where the shift will happen. Instead of following the traditional approach of collecting information separately, manufacturers will be able to bring it together and use it more effectively.
However, I believe humans will continue to play a role in the process. AI should not dominate humans, and nobody would want that. That is how I see the future of AI in manufacturing.