AI Agent Operational Lift for Starplus Energy in Kokomo, Indiana

Why AI matters at this scale

StarPlus Energy, a joint venture between Stellantis and Samsung SDI, sits at the heart of America's EV transition. As a mid-market manufacturer with 201-500 employees and a brand-new facility in Kokomo, Indiana, the company faces a unique pressure: it must rapidly scale production of highly complex lithium-ion battery cells while maintaining the quality and safety standards of its global parent companies. At this size, StarPlus is large enough to generate meaningful production data but likely lacks the massive digital infrastructure of a Tier 1 automotive giant. This makes it an ideal candidate for pragmatic, high-ROI AI adoption that can leapfrog traditional automation.

The Quality-Yield Imperative

The single most impactful AI opportunity is in computer vision for inline defect detection. Battery electrode manufacturing involves coating foils with slurries at high speeds. Microscopic defects—pinholes, agglomerates, or uneven coating—can lead to internal short circuits and thermal runaway. Deploying deep learning models on high-speed camera systems can catch these defects in real-time, allowing for immediate correction. For a plant targeting gigawatt-hour scale output, improving yield by even 2-3% through reduced scrap translates to millions in annual savings and, critically, prevents safety recalls.

Beyond the Production Line

Two additional opportunities offer strong ROI. First, predictive maintenance on mixing and coating equipment. These are bottleneck assets; unplanned downtime cascades through the entire plant. Vibration and temperature sensors feeding an ML model can forecast bearing failures or pump degradation weeks in advance, enabling condition-based maintenance. Second, AI-driven energy optimization for the formation cycling process. This multi-day step charges and discharges cells for the first time and is enormously energy-intensive. A reinforcement learning agent can dynamically schedule cycling to leverage off-peak electricity pricing and optimize thermal management, potentially cutting energy costs by 10-15%.

Navigating Deployment Risks

For a 201-500 person firm, the primary risk is not technology but execution. A 'data lake' project without a clear use case will fail. The pragmatic path is to start with a vendor solution for visual inspection that runs at the edge, avoiding complex IT integration. Workforce upskilling is equally vital; line operators must trust the AI's recommendations, not see it as a threat. Finally, cybersecurity for operational technology must be foundational, as connecting production lines to analytics platforms creates new attack surfaces. By focusing on these concrete, high-value use cases, StarPlus Energy can build a data-driven culture that directly supports its mission of delivering safe, affordable EV batteries at scale.