Request a demo
Services

Built on your tags, not a rip-and-replace platform

Every engine below runs against your existing PLC/OPC UA data. Pick one, or run all three as a single monitoring stack.

How it's packaged

Three tiers, one set of tags underneath

Each tier adds capability on the same live data — nothing is rebuilt or replatformed to move up a tier, and nothing writes to your PLC until you're ready for it to.

Tier 1 Read-only

Visibility

Live control charts and process capability, checked before they're shown, not after.

  • Xbar-MR charts, Cp/Cpk, Pp/Ppk, sigma drift
  • Every value passes a validation check first — sample size, autocorrelation, sensor faults
  • Data that doesn't clear validation is flagged, not smoothed over or hidden
  • Nothing writes back to the PLC
Tier 2 Read-only

Insight

Everything in Visibility, plus the modeling layer that tells you why a number moved.

  • Influence scoring across an ensemble of methods — Pearson, Spearman, PLS, Ridge
  • EWMA/CUSUM anomaly detection surfaced as early trend, not a raw alarm
  • Validation gates the model directly — unstable input suppresses or flags the score, it doesn't get rounded off and shown anyway
  • Still read-only
Tier 3 Writes to PLC

Control

Everything above, plus validated output acting on your process — alarms, interlocks, discrete control.

  • Requires a track record on Tiers 1–2 against your own data first — this isn't switched on day one
  • If validation fails at runtime, the system holds. It doesn't guess
  • Priced and contracted separately, with its own commissioning and testing
On validation: every chart and score above has already been checked for enough samples, for autocorrelation, for sensor faults — before you ever see it. That's not a footnote feature. It's the reason Tier 3 is allowed to exist at all: nothing gets to act on your process until it's been honest about its own uncertainty for long enough to earn it.
Under the hood

What's actually running in each tier

The six engines below are the building blocks — Tier 1 runs the first three, Tier 2 adds influence scoring, Tier 3 turns validated output into action.

TAG: SPC_ENGINE

Statistical process control

Xbar-MR charts, Cp/Cpk, Pp/Ppk, sigma drift, and out-of-spec counts — computed from sample statistics, not the population-variance shortcut that quietly inflates confidence.

TAG: ANOMALY_DETECT

Real-time anomaly detection

EWMA and CUSUM with time-varying control limits. We account for autocorrelation in process data rather than trusting ordinary cross-validation, which silently breaks on time series.

TAG: INFLUENCE_SCORE

Influence & root-cause scoring

Ranks which of your predictor tags most affects a response variable, using an ensemble of methods rather than trusting any single correlation metric.

TAG: OPERATOR_DASH

Operator dashboards

Status-first views — color-coded tiles that show "fine" vs. "needs attention" at a glance, with root-cause detail one click away instead of competing for attention up front.

TAG: LIVE_STREAM

Live streaming architecture

A fast poller filling a buffer and a slower scorer reading snapshots from it — so live scoring never blocks on live data collection.

TAG: INTEGRATION

Ignition / SCADA integration

Direct integration with Ignition Perspective and similar SCADA layers, including dynamic tag binding so one view scales across many assets.

Not sure which engine fits your process?

Tell us what you're monitoring today and we'll tell you what's realistic to add.

Request a demo