A roughly 30% increase in total component placements across four manufacturing sites — that is the headline number HARMAN Automotive reports from an artificial intelligence deployment now running across its global electronics manufacturing network.

The automotive electronics supplier, a Samsung subsidiary headquartered in Palo Alto, Calif., deployed Arch Systems' Factory Intelligence Platform across its plants, starting with its facility in Suzhou, China. According to Arch Systems — not HARMAN's own production data — the deployment has delivered a 37% improvement in defects per million units and a 6- to 8-percentage-point improvement in overall equipment effectiveness across multiple sites.

Those figures come from the vendor, and suppliers' AI productivity claims routinely warrant scrutiny against actual line-level output. HARMAN has not independently published the underlying production data. What is confirmed: HARMAN rolled out the platform at Suzhou first, then extended it across a four-site electronics manufacturing network.

"For us, AI is an enabler to drive quality and productivity across our manufacturing sites, while improving transparency and creating greater synergies across our global operations," said Don Barnes, vice president of manufacturing excellence at HARMAN Automotive.

The platform works by combining data streams from three layers of the factory: equipment on the shop floor, manufacturing execution systems and enterprise systems. AI agents then use the consolidated data to investigate production issues spanning quality, planning and operations. In practice, that shifts root-cause investigation from manual correlation across disconnected systems to agent-driven queries against a unified dataset.

The architecture matters for Tier 1 electronics suppliers under cost pressure. HARMAN builds cockpit electronics, telematics and audio systems for multiple OEMs, and placement throughput and defect rates on surface-mount technology lines directly drive per-unit cost. A 37% DPM improvement, if sustained across the network, would compound across the millions of units HARMAN ships annually.

FORVIA HELLA is testing the same technology. The German Tier 1 has begun a pilot on its surface-mount technology lines using real-time downtime labeling and AI-assisted root-cause analysis to identify production problems. The pilot stage means no capacity or quality claims yet — only stated intent to scale where results justify it.

"At FORVIA HELLA, we are focused on scaling AI where it creates measurable business impact," said Sven Hoenecke, president and CEO of FORVIA HELLA Electronics, Americas.

Hoenecke's phrasing signals the qualifier embedded in both programs: measurable impact. HARMAN's numbers cover four sites out of its broader global network, and the vendor-reported basis leaves open how the gains translate into confirmed capacity or headcount outcomes. Neither company has disclosed the investment size of the deployments or a timeline for extending the platform to additional plants.

What to watch next: whether FORVIA HELLA converts its SMT pilot into a production rollout with its own verified metrics, whether HARMAN extends the platform beyond the current four sites, and whether either supplier publishes independently audited OEE data to back the vendor's claims.