Industry · Aquaculture
Your salmon operation's fingerprint knows when something is wrong. Before any alarm fires.
Water chemistry shifts. FCR drift. Equipment deviation. These signals appear in your time-series data hours before any operational impact is visible. Eyer learns to read them from your historical data.
Evidenced result: land-based salmon producer, Norway
Eyer ran on historical water chemistry data from a land-based salmon producer. The anomaly pattern appeared 2–3 hours before any operational impact was visible. No threshold was ever breached. The existing alarm system never fired. Eyer saw it.
Anonymised. Results from running Eyer on historical customer data.
The aquaculture challenge
High failure consequence. Dense sensor data. Dynamic baselines.
Consequence severity
A water chemistry deviation that goes undetected for 2–3 hours means biomass loss, regulatory reporting, and emergency intervention costs that dwarf any monitoring spend.
Dynamic baselines
Temperature, oxygen, CO2, salinity, pH and turbidity all shift with season, fish age, stocking density, and feeding cycle. Static thresholds cannot track these dynamics.
Regulatory pressure
Aquaculture compliance requirements are tightening across markets. Lenders and insurers want operational records, not assurances. Eyer produces a timestamped audit trail per sensor, per event.
How Eyer applies to aquaculture
Fingerprint learning on water chemistry, feeding, and equipment.
Eyer handles any time-series data from RAS controllers, OT equipment, or IoT sensors. Data from your tanks, feeding systems, and pumps reaches Eyer via Telegraf, Prometheus, or a direct push. No new sensors. Eyer learns the normal behaviour of each system from historical data. Anomalies surface early, before operational visibility.