India has no shortage of “smart factory” decks. But most of them are still stuck at sensors, dashboards, and a few pilot lines. Apollo Tyres is one of the rare manufacturers that has moved beyond presentations and into production-scale AI, compressing tasks that once took hours into minutes and turning machine data into a live decision system on the shopfloor.
From two hours to ten minutes
In one of Apollo’s flagship plants, diagnosing a critical machine issue used to be a slow, expert-heavy exercise. Engineers had to pull historical logs, inspect trends across SCADA, PLC, and MES systems, and often walk the line to piece together what was actually happening. That process could easily run to 90–120 minutes, during which a key machine might be idle or running below capacity.
Today, much of that work is handled by an agentic AI layer built on top of the company’s existing IoT and data infrastructure. Instead of manually stitching together data from multiple systems, engineers pose natural-language questions to an AI “reasoner” that has real-time access to thousands of machine tags, historical patterns, and maintenance records. The AI can surface likely root causes, suggest parameter adjustments, and even trigger workflows in the plant’s maintenance systems. What once took nearly two hours can now be narrowed down in around ten minutes, with far fewer trial-and-error interventions.
Building on years of digital groundwork
Apollo’s AI moment did not appear out of nowhere. Over the last several years, the company has invested in industrial IoT, plant connectivity, and cloud-based ERP and manufacturing systems, including large-scale rollouts on SAP’s cloud stack. Those programmes created the raw ingredients: cleaner data from machines, more consistent processes across plants, and a single source of truth for production, quality, and inventory.
Earlier initiatives focused on making factories “smarter” with condition monitoring, Overall Equipment Effectiveness (OEE) tracking, and better visibility into the supply chain—from raw materials to finished tyres leaving the dock. The move to AI is essentially the second act: shifting from visibility to autonomy, from dashboards that tell you what happened to systems that can recommend (and in some cases initiate) what should happen next.
How agentic AI sits on the shopfloor
At Apollo, AI is not a black box in the cloud. It sits in the flow of daily operations. IoT gateways stream machine data into a central platform, where specialised models are trained to spot patterns: vibration signatures that precede bearing failure, temperature profiles that correlate with surface defects, or pressure variations that lead to premature wear.
On top of this, an agentic AI layer acts as a conductor. It can:
- Correlate anomalies across multiple machines and lines.
- Compare current behaviour with historical “golden runs”.
- Suggest optimal setpoints or maintenance windows.
- Open tickets in maintenance and update shift reports automatically.
The interface is intentionally simple. Line engineers and supervisors interact with the system via dashboards and natural language prompts rather than code or complex query builders. That design choice is critical in a plant environment, where decisions are time-sensitive and digital literacy levels vary.
Tangible gains: downtime, scrap, and energy
Public case material around Apollo’s AI deployments highlights concrete operational wins. Diagnostic time reductions from around two hours to about ten minutes for specific classes of machine issues translate directly into less unplanned downtime and higher throughput. Earlier AI and ML deployments have also been credited with measurable cost savings, with one report citing around ₹1 crore saved through better process optimisation and defect reduction.
Beyond pure uptime, AI-driven insights help tighten process windows, improving consistency and reducing scrap. Even small improvements in error rates can have outsized impact in tyre manufacturing, where margins are thin and volumes are high. Energy is another frontier by analysing load curves and machine utilisation patterns, AI can recommend shifts in scheduling and setpoints that trim energy consumption without hurting output.
Changing the role of people, not removing them
A common fear on any factory floor is that AI will replace people. Apollo’s experience so far suggests a more nuanced shift: roles are changing, not disappearing. The company’s own leaders have publicly emphasised reskilling and human-in-the-loop control as central to the transformation.
Maintenance engineers who once spent most of their time chasing recurring issues are now spending more time on reliability engineering and continuous improvement, using AI insights as a starting point rather than an answer. Line operators are being trained to interpret AI recommendations, override them when necessary, and feed back real-world context creating a loop where the system and the humans get better together.
This rebalancing is echoed across Indian manufacturing CIO circles, where the conversation has moved from “automation versus jobs” to “how do we make operators 5–10x more effective with AI tools?”.
Lessons for the rest of Indian manufacturing
Apollo’s journey offers a few clear takeaways for other Indian manufacturers watching from the sidelines:
- AI is a second‑order investment. The company could move quickly on agentic AI precisely because it had already done the heavy lifting on IoT, connectivity, and data platforms. Plants still wrestling with siloed PLCs and manual logs will struggle to leapfrog straight into advanced AI.
- Start with high-friction use cases. Apollo began in areas where the pain was obvious: long diagnostic cycles, recurring quality issues, and process bottlenecks that directly hit throughput and cost. Solving those built confidence and a business case for broader rollout.
- Keep the interface human. Fancy models don’t help if operators can’t or won’t use them. By fronting its AI with intuitive dashboards and question-driven interfaces, Apollo made the technology feel like an assistant, not an audit.
- Partner where it matters. From cloud platforms to AI specialists, Apollo tapped external partners instead of building everything in-house. That allowed its internal teams to focus on domain expertise tyre processes, plant operations, and quality.
A glimpse of where Indian factories are heading
Apollo Tyres is not the only Indian manufacturer experimenting with AI, but it is one of the clearer examples of what a genuine “factory of the future” looks like when you move beyond PowerPoint. Machines stream data by default. AI agents watch in real time, helping engineers intervene earlier and smarter. Processes are nudged continuously toward more uptime, less waste, and lower energy use.
For the broader ecosystem auto components, steel, chemicals, food processing the message is simple. The path to AI in manufacturing doesn’t begin with moonshot robots; it begins with better questions: Where are we losing time? Where do experts get stuck in repetitive analysis? Where is our data trapped? Once those are answered, the kind of transformation Apollo is demonstrating stops looking “insane” and starts looking inevitable.
