Manufacturing has entered an era where the factory floor is as much a data environment as it is a production environment. Few industries illustrate this shift more clearly than precision bearing manufacturing, where microscopic tolerances, high-volume output, and rigorous quality demands leave little room for error — and enormous room for data-driven optimization.
At NBC Bearings, one of India’s leading bearing manufacturers, that shift is being driven from the top of the technology organization. Dr. Lokesh Agrawal, CTO of NBC Bearings, is leading an effort to fuse decades of mechanical engineering expertise with AI, digital twins, IoT-enabled sensing, and advanced analytics — not as a bolt-on IT initiative, but as a redefinition of how the company designs, builds, and monitors its products.
“At NBC Bearings, we view digital transformation not as an IT-led exercise, but as a business and engineering transformation,” Dr. Agrawal says. “Our objective is to use technology to improve productivity, quality, flexibility and speed, while creating new avenues for product and process innovation.”
Four Pillars, One Outcome-Driven Strategy
Rather than pursuing digital initiatives in isolation, NBC Bearings has organized its transformation around four interconnected pillars: smart manufacturing, digital engineering, intelligent products, and a digital supply chain.
“Smart manufacturing” means deepening the use of automation, connected equipment, advanced analytics and machine learning “to improve product and manufacturing performance and enable faster decision-making,” Dr. Agrawal explains. “Digital engineering” involves weaving simulation, sensor-based technologies and digital models directly into product development. Perhaps most striking is the company’s push toward “intelligent products” — a move, in Dr. Agrawal’s words, “from conventional mechanical components towards intelligent bearing solutions that can generate data, enable condition monitoring and support prognostic capabilities.”
Underpinning all four pillars is a strict investment discipline. “The underlying principle is that every technology investment must have a clear business outcome, whether that is improving quality, reducing cycle time, increasing asset utilisation, improving energy efficiency or enabling a differentiated product,” he says.
AI as an Augmentation Layer, Not a Black Box
For Dr. Agrawal, AI’s greatest value in a high-volume manufacturing environment lies in decision support rather than autonomous decision-making. “Our approach is to use AI where it can augment engineering and operational decision-making,” he says, pointing to equipment condition monitoring, predictive maintenance, process parameter optimisation, quality analytics and production planning as the current focus areas.
The ambition goes further still. “Instead of responding only after a deviation or failure, the objective is to identify patterns in process and equipment data early and move towards predictive and eventually prescriptive decision-making,” he notes — describing a broader industry arc from diagnostics to prognostics that NBC Bearings is actively building toward.
But Dr. Agrawal is candid that algorithms alone won’t get manufacturers there. “AI adoption in manufacturing cannot be viewed simply as deploying algorithms,” he cautions. “The quality of the underlying data, process discipline, domain knowledge and ability to integrate insights into hierarchical decision-making are equally important.”
Digital Twins Meet Traditional Engineering
Nowhere is the fusion of old and new more visible than in NBC Bearings’ R&D process. The company is combining decades of mechanical design expertise with simulation, sensor data and digital modeling to compress development cycles. “Our traditional engineering knowledge has been digitalized with AI capability providing a first concept which meets various criterion,” Dr. Agrawal says. “Simulation allows us to further optimize design before physical prototypes are built, thereby improving the efficiency of the development cycle.”
Sensor-enabled bearings are also feeding operational data back into engineering models — a capability with direct implications for material efficiency. “Sensor-enabled bearings can provide operational data that can be fed into digital models, helping us understand residue life and optimise designs without relying entirely on conservative safety factors,” he explains. “This has direct relevance to lightweighting and material optimisation.”
Importantly, Dr. Agrawal is clear that digital tools are additive, not a replacement for physical testing. “The objective is not to replace physical validation,” he says. “Rather, digital tools allow us to make the design process more informed, reduce the number of iterations and focus physical testing where it adds the most value.”
Engineering for the Electric Vehicle Era
As automakers accelerate the shift to electric powertrains, bearing manufacturers face a fundamentally different set of engineering constraints — higher speeds, different load profiles, and stricter NVH (noise, vibration, harshness) requirements. “Electrification changes the operating environment for bearings significantly,” Dr. Agrawal says. “Efficiency becomes even more critical because every source of energy loss directly affects overall vehicle efficiency and range.”
NBC Bearings has been preparing for this shift for years, developing “solutions around low friction, high-speed operation, lightweighting, sensor integration and improved durability,” he says. That includes low-torque bearing designs engineered specifically to cut energy losses, as well as component-level lightweighting through non-metallic and polyamide cages. “Lower cage mass can reduce centrifugal forces and friction, particularly in high-speed applications,” he notes.
Sensor integration is central to the company’s EV strategy as well. “The bearing is evolving from being purely a mechanical component to becoming a component capable of generating information about operating conditions,” Dr. Agrawal says — a capability that supports condition monitoring, prognostics, and ultimately smarter vehicles.
The bigger picture, he says, isn’t just about unit volume. “We are therefore looking at EVs not simply as a change in the number of bearings used, but as an opportunity to increase the technological content and value delivered by each bearing.”
Redefining the Smart Factory
Ask Dr. Agrawal what a “smart factory” actually means, and the answer isn’t about robot density. “For us, a smart factory is not defined by the amount of automation installed,” he says. “It is defined by how effectively data from machines, processes and people is converted into better decisions.”
That vision relies on a broad technology stack — “automation, industrial IoT, sensors, data analytics, machine learning, digital dashboards and condition-monitoring systems” — aimed at pushing the organization from reactive to predictive, and eventually prescriptive, manufacturing.
Technology, however, only goes as far as the people using it. “A smart factory requires engineers and operators who understand both the manufacturing process and the digital tools supporting it,” Dr. Agrawal says. “Reskilling therefore becomes an important component of Industry 4.0 adoption.”
Data Governance as the Foundation, Not an Afterthought
With more machines, sensors and quality systems generating data across NBC Bearings’ operations, Dr. Agrawal is emphatic that raw data volume is not the goal. “The value of data comes from its quality, context and usability — not simply from the volume generated,” he says.
That data is already proving its worth in root-cause analysis and predictive maintenance, and it’s opening a new frontier at the product level. “Our move towards intelligent solutions also creates an important product-level data opportunity,” he says. “Real-time operating information can support condition monitoring and prognostics, while feeding insights back into future product development.”
None of it works without governance. “If data is inconsistent, poorly structured or not properly governed, even sophisticated AI models will produce limited value,” Dr. Agrawal warns. “We therefore need clear ownership of data, standardised definitions, appropriate access controls, cybersecurity measures and processes to ensure data quality.” The end goal, he says, is “a trusted data environment where engineers and business teams can use information confidently to make faster decisions.”
Looking Ahead: Convergence, Not a Single Silver Bullet
Asked which emerging technology will matter most over the next five years, Dr. Agrawal resists naming just one. “I believe the biggest impact will come not from any single technology, but from the convergence of AI, advanced materials, robotics, connected products and digital engineering,” he says.
He anticipates AI’s role expanding well beyond analytics. “AI will increasingly move from being an analytical tool to becoming an engineering and operational co-pilot with human in loop all the time,” he says — a distinction he considers essential. “While I see AI advancing, keeping human in loop is essential to keep the innovation and solution centricity focus.” Agentic AI, he suggests, may eventually help stitch together capabilities across the manufacturing and engineering workflow, though always with people in the decision loop.
On the factory floor, he expects robotics and autonomous systems to take on more repetitive, high-precision and ergonomically demanding tasks, often paired with machine vision for quality inspection. Materials science remains just as critical to the roadmap. “Lightweighting is not simply about removing material,” he says. “It requires the right combination of geometry, material selection, heat treatment and manufacturing processes to achieve the required durability and performance for application specific needs.”
And the humble bearing itself, in his telling, is on a trajectory from passive component to active data source. “Intelligent bearings will increasingly connect the physical and digital worlds,” he says. “Condition monitoring and digital twins can enable bearings to move from being passive components towards intelligent components capable of providing actionable information about their own operating condition.”
For Dr. Agrawal, execution will separate the winners from the rest. NBC Bearings’ preparation, he says, rests on three commitments: “building technology capabilities, strengthening our R&D and engineering ecosystem, and developing the digital skills of our people.” As he puts it, summing up the philosophy behind the entire transformation: “The organisations that will benefit most from emerging technologies will be those that can combine technology with deep domain expertise and strong execution.”