manufacturing · European frozen food manufacturer
before process deviation impacts product quality
A frozen food manufacturer applied Eyer to cold chain and processing data — detecting quality-affecting process deviations before they reached specification limits. No new sensors required.
Each metric tracked with lower, central, and upper learned baselines
Process anomalies detected before downstream product quality is affected
The operation
A European frozen food manufacturer with multiple processing lines handling vegetables, fruit, and herbs. Continuous production with strict cold chain requirements, food safety certification obligations, and regulatory traceability requirements. Existing SCADA-based monitoring focused on compliance threshold adherence.
The challenge
In food manufacturing, process deviations that are not caught early have two consequences: product quality issues that result in batch rejections or customer complaints, and food safety compliance events that trigger regulatory reporting and potential production line shutdowns.
The existing alarm system was configured for compliance — it would fire when a parameter crossed a regulatory or quality specification limit. It was not configured to detect the developing process anomalies that precede specification limit crossings by hours, at which point corrective action is still straightforward.
What Eyer found
Eyer was run on the historical process data from two production lines as a Fast Forward analysis. The analysis identified anomaly patterns across cold chain temperature management, blanching and cooling process parameters, and freezing line performance metrics.
The most significant finding was a recurring pattern in the cold chain temperature data: a gradual drift from learned baseline behaviour that, when it appeared, consistently preceded temperature excursion events that triggered quality holds. The drift was statistically detectable in the data several hours before any threshold was crossed. The existing system had no way to see it.
Correlation across the production line
Eyer's correlation analysis identified a relationship between the cold chain temperature drift pattern and upstream blanching process parameters. The temperature anomaly was not developing in isolation — it was consistently correlated with specific operating states in the upstream process that could be identified and adjusted.
This correlation was invisible to the existing alarm system, which treated each threshold crossing as an independent event. It was visible in Eyer's correlation model, which maps the statistical relationships between signals as they actually exist in the operational data.
Food safety and regulatory traceability
Food manufacturing carries specific regulatory traceability requirements. The operational audit trail that Eyer produces — timestamped, asset-level, capturing both normal operation and deviation patterns — provides the structured operational record that food safety auditors and certification bodies require. This is a secondary value stream that sits alongside the early detection capability: not just catching problems before they happen, but proving you have the operational intelligence infrastructure in place.
See what Eyer finds in your operation
Run Eyer on your historical data.
First findings within one week. No new sensors. No infrastructure changes.
Book a Fast Forward