Manufacturing / Automotive electronics
HARMAN: Factory operations alerts and defect root cause investigation
- Company
- HARMAN
- Country
- Multiple countries
- Adoption stage
- In operation
- Source published
- Date basis
- The date the source was published. It can differ from the date adoption started.
- How the source was checked
- Read the full source text
The work problem
As HARMAN expanded automotive electronics production worldwide, keeping operational performance consistent across plants became increasingly difficult. Manufacturing expertise was concentrated in a small group of engineers, and specialists frequently traveled between plants to diagnose issues. Each plant used its own locally built tools and methods, so successful improvements were hard to transfer to other plants. Dashboards showed problems but did not consistently lead to standardized action, and workforce turnover put knowledge at risk.
Technology and data
HARMAN deployed Arch Systems' Arch Factory Intelligence Platform across its production network. Starting with a lighthouse deployment at its plant in Suzhou, China, it connected real-time machine and production data with prioritized alerts, guided workflows and role-based actions. Operators, engineers and production leaders use these alerts and guidance to respond to shop floor issues. Control rooms at connected plants show production performance and the execution of improvement activities in real time. Capabilities validated in Suzhou are being standardized and extended to other plants. Since the Automotive News PACE Award site visit, the two companies have continued to advance the AI agent capabilities. Starting from a defect found by automated optical inspection (AOI) or an in-circuit test (ICT) failure, an AI agent traces the issue upstream into the component placement process and analyzes material and equipment history. It then tests competing causes against the evidence to narrow down the root cause and recommends a specific corrective action. Arch says this reduces the investigation steps in which engineers manually gathered and analyzed evidence across machines and systems.
Results
According to the Arch Systems case study, total component placements across four manufacturing sites grew by about 30%. The same source reports a 37% improvement in defects per million units (DPMU) and a 6 to 8 percentage point rise in overall equipment effectiveness (OEE) across multiple sites. Arch wrote that HARMAN made better use of existing manufacturing capacity without adding new production lines. HARMAN's global head of automotive manufacturing said site-by-site improvements had become a capability that can scale across the global manufacturing network. The initiative was named a finalist for the 2026 Automotive News PACE Award.
Limits and open questions
The figures come from a case study published by Arch Systems and do not state the measurement period, the baseline or the names of the four sites. The AI agent root cause capability is described only through expected benefits, and the share of the results attributable to AI is not reported.
Sources
Compiled from public sources. These are not results from ATF Works customers.