Manufacturing Data Trapped in Proprietary Silos Blocks Smart Manufacturing Adoption by Small and Medium Manufacturers

Smart manufacturing — using sensors, analytics, and AI to optimize production in real time — could reduce U.S. manufacturing energy consumption by 5–7% and CO2 emissions by 7% per year. But the 250,000+ small and medium-sized manufacturers (SMMs) that form the backbone of U.S. manufacturing cannot adopt these technologies because manufacturing data is trapped inside proprietary software systems that don't communicate with each other. Each equipment vendor, process control system, and enterprise software tool stores data in its own format behind its own walls. A National Academies study commissioned by Congress found that U.S. manufacturing is "organized for vertical optimization, transactional interaction, and data protectionism and isolation, not interoperability." Without cross-vendor data exchange, an SMM cannot aggregate data from its own equipment — let alone benchmark against industry peers or access shared analytics.

The U.S. has over 250,000 SMMs employing millions of workers and producing the majority of manufactured goods. Large manufacturers with dedicated IT departments can build custom integrations, but SMMs typically lack capital, technical staff, and time to do so. The result is a two-tier system where smart manufacturing benefits concentrate among large firms while the economic majority of manufacturers falls further behind. CESMII projects that smart manufacturing adoption averages only 9–10% market penetration over the next decade without intervention. The National Academies report proposed a "Cyber Interstate" — analogous to the highway system — as a national data infrastructure for manufacturing, but no such infrastructure exists. Meanwhile, 61% of small manufacturers have already experienced cyberattacks, and 35% say cybersecurity vulnerabilities further inhibit their adoption of connected technologies.

Existing standards (OPC-UA, MTConnect) provide protocols for machine-to-machine communication but don't solve the semantic interoperability problem — different vendors use different data models, naming conventions, and contextual metadata for the same physical quantities. Manufacturing integration middleware exists but requires expensive customization for each deployment. The Manufacturing Extension Partnership (MEP) provides technical assistance to SMMs but lacks smart manufacturing focus and funding. Manufacturing USA institutes (CESMII for smart manufacturing, CyManII for cybersecurity) develop reference architectures and training but cannot mandate industry adoption. Cloud-based manufacturing platforms (e.g., from Siemens, GE, PTC) offer integration but lock manufacturers into a new form of vendor dependency. The fundamental barrier is that no economic incentive exists for equipment vendors to make their data formats open and interoperable — proprietary data creates customer lock-in.

Open, vendor-neutral data standards and semantic models for common manufacturing processes that SMMs can adopt without custom engineering. A shared data infrastructure — the report's "Cyber Interstate" concept — that provides secure data exchange, benchmarking, and analytics as a public utility rather than a premium service. Cybersecurity frameworks specifically designed for resource-constrained manufacturers (not adapted from enterprise IT). Critically, the incentive structure needs to change: either through regulatory requirements for data portability, industry consortium agreements, or government-funded shared platforms that bypass proprietary lock-in.

A team could select a common SMM manufacturing process (e.g., CNC machining, injection molding, or 3D printing) and build a proof-of-concept data translator that maps between 2-3 vendors' proprietary data formats into a common schema, demonstrating what information is lost or gained in translation. Alternatively, a team could design and prototype a lightweight, low-cost sensor-to-cloud pipeline for a single machine type using open protocols, targeting the specific constraints of SMMs (limited IT staff, legacy equipment, minimal budget). Skills needed: data engineering, IoT/embedded systems, and basic manufacturing domain knowledge.