FoundryNet / Solutions / Mixed fleet normalization

MIXED FLEET · CROSS-VENDOR NORMALIZATION

Your AI Can't Reason Across a Mixed Fleet Without Normalization

16,908 confirmed mappings·18 OEM families·366 canonical fields·30-second setup

One machine is an integration. Five hundred machines from eight manufacturers is a translation problem. Forge collapses every OEM dialect into one schema your agent already understands.

Get Forge on AWS Marketplace →See pricing →

Paste telemetry from any two machines and see them come out as the same schema. No POC required.


The problem

A single machine is readable. A fleet is not.

A frontier model can read one machine's raw telemetry unaided. Point it at a Fanuc dump and it will work out that SpindleSpeed is an RPM. That is not the hard part, and pretending otherwise insults the reader.

The hard part starts at machine number two. Eight OEMs on one floor means eight naming conventions, three unit systems and no shared notion of what "load" means. Spindle load is a percentage on a Haas, an amperage on an older Okuma, and a torque ratio on a Siemens Sinumerik. An agent asked to rank your worst-performing machines is now comparing percentages against amps and calling the answer analysis.

Roughly 70% of installed manufacturing equipment is 20+ years old, so the fleet you actually own is a museum of naming decisions made by people who never expected an LLM to read them. And 58% of industrial AI project resources go to integration before a model touches a single machine — which is a large part of why 40% of agentic AI projects will be cancelled by 2027 and 88% of AI pilots never reach production.

70%Equipment 20+ years old
58%Of AI budget spent on integration
88%Of AI pilots fail
15-20%Utilization gain, single platform

What normalization actually is

Same physical quantity. Same field name. Every time.

Forge maps raw vendor tags onto 366 canonical fields. Four machines, four vendors, four dialects, one output shape:

Raw tag inCanonical field outWhere it came from
haas   Q600_1: 3400spindle_speed_rpm: 3400Next Gen Control Q-variable
fanuc   SpindleSpeed: 3400spindle_speed_rpm: 3400FOCAS / MTConnect stream
siemens   DB21.DBD40: 3400spindle_speed_rpm: 3400S7 data block offset
mazak   S_ACT: 3400spindle_speed_rpm: 3400SmoothX actual spindle

Your agent writes one query against spindle_speed_rpm instead of four adapters against four vendors. That is the entire value proposition, and it only shows up at fleet scale.


How it holds up

Five layers in, six validators out

Five-layer normalization

Curated corpus → embedding match → LLM research → physics validator → signal classifier. Each layer only sees what the one above it could not resolve, so the common case never pays for the hard case.

Six read-time validators

Rate-of-change, stuck sensor, dropout, operating mode, correlation and confidence decay. A frozen sensor reading a plausible number is caught before your agent acts on it.

Self-healing corpus

A tag resolved once by the LLM is cached, confirmed after 5 uses, auto-packed at 10, and served from Layer 1 forever after. Novel-tag cost is per new tag, not per call — it trends to zero as the fleet stabilises.

Abstention over guessing

Fields below the confidence floor return unmapped rather than force-fitted. An agent acting on a confidently wrong mapping is worse than an agent that knows it is missing a field.


30-second deployment

Add the whole fleet with one call

No SDK, no per-tag mapping table, no historian project. Same call for every manufacturer — only the oem hint changes, and it is optional:

claude mcp add --transport http forge https://mcp.foundrynet.io/mcp --header 'Authorization: Bearer YOUR_API_KEY'
# one endpoint, any manufacturer
curl -s https://forge.foundrynet.io/v1/normalize \
  -H 'Content-Type: application/json' \
  -d '{"data": {"S_ACT": 3400, "spd_load": 85}, "oem": "mazak"}'
{"spindle_speed_rpm": 3400, "spindle_load_pct": 85, "coverage": "100%", "confidence": 0.97}

Unknown manufacturers still normalize. The vertical gate disables itself when the OEM is unrecognised, so you get clean fields on day one and higher-confidence mappings once you pass oem and model.

Get Forge on AWS Marketplace →See pricing →

Related

Keep reading

Legacy protocol integration

Most of that mixed fleet is 20+ years old. 14 protocols, RS-232 to OPC UA.

The missing data layer

Why the pilot failed at the data layer and not the model.

Cross-vendor equipment intelligence

The full OEM grid: 18 families, 55 manufacturer pages.


From the newsletter

The Machine Agent

Longer-form writing on industrial AI infrastructure by Foundry Labs.

The Tower of Babel

The most expensive unsolved problem in industrial AI

Read on Substack →