SEMMAPPTelemetry module
CONCEPT
AI analytics · mockup

What the plant is telling you, and how sure we are

Every number on this page is calculated in code from the asset register, the readings and the telemetry. Words are drafted around those numbers, never the other way round. Each insight says how mature it is: calculated today, waiting for a season of telemetry, or machine learning that needs about three seasons of this factory's own history.

01Data maturity

Manual readings and lab results in SEMMAPP Continuous telemetry Machine learning threshold, about 3 seasons

02Insight feed

Ranked by severity, then by how soon the trend reaches a limit. Projections are straight line fits on measured readings with a range from the slope's standard error. That is calculation, not machine learning.

03Loss tree

Top assets by hours lost
Cumulative share of all downtime, Pareto order

04Correlations explorer

Pick a pair. Telemetry series that have no history yet use a deterministic synthetic season so the view can be judged; those are labelled. Correlation shows what moves together, it does not prove cause.

05Energy

Steam on cane, power per ton and boiler efficiency, calculated from the live tag values (design values in this mockup). The calculation is shown in full so an engineer can check it.

Boiler 1 efficiency, indirect methodCalculated, available now
Efficiency against flue gas oxygen

06Ask the plant Mockup

The assistant answers from the same calculations as the rest of this page and cites the tags and channels it used. It declines market and price questions, and it abstains when a question needs history the factory does not have yet.

Try asking

07Unlocks later