Remaining Useful Life Without Run-to-Failure Data
You don't need a graveyard of failed bearings to forecast a repair, and the number on the dashboard was never the point.
Field Notes · Author
Andrus Nõmm
Andrus Nõmm leads Zoniax OÜ in Tallinn, Estonia, where the team instruments processing plants with ruggedized sensors, edge telemetry, and learning models.
AI on Both Sides of the OT Firewall
Brussels just put AI on both sides of the plant firewall. Read from the control room, the EU's new Cyber-AI Action Plan is part threat model, part toolkit.
What to Measure Before You Electrify the Boiler
A heat pump's economics live or die on two signals the boiler house has never logged: the temperature each duty truly needs, and the shape of the load across the day.
The Model That Goes Quiet: MLOps on the Plant Floor
Deployment is day one for an industrial model, not the finish line: how drift, retraining, and ownership decide whether it survives its second year.
Scoring Your CRA Readiness From the Plant Floor
ENISA just shipped a five-domain maturity model for the Cyber Resilience Act, and read against a real processing line it works less like a scorecard than a gap list.
The AI Act Slipped. The Machinery Deadline Didn't.
The digital omnibus pushed high-risk AI obligations out to December 2027, but the machinery regulation still lands in January 2027, and that gap is where the work sits.
What Model Predictive Control Actually Buys a Plant
Five claims plant teams hear about MPC, from vendors and skeptics alike, checked against the control-engineering record.
IED 2.0: The Meters You Need Before July 2027
The revised Industrial Emissions Directive asks large plants to prove their water, energy and materials performance, and the number it wants is a ratio most sites can't yet defend.
The Cyber Resilience Act's 24-Hour Clock at the Edge
From 11 September 2026, the maker of every connected industrial device owes Europe an early warning within 24 hours of an exploited flaw - and the plant is usually where that flaw is seen first.
How to Measure ROI on Industrial AI
A plant-floor method for baselining, attributing, and pricing industrial AI value before the dashboards outrun the returns.
From Anomaly to Root Cause in Continuous Plants
An anomaly score tells you something changed. Getting to what changed first is a separate, harder problem in correlation, direction, and disciplined alarms.
Running LLMs at the Edge Inside the Plant
A field note on putting a language model in the rack: the hardware, the memory-bandwidth wall, the heat, and why the box never closes a control loop.
ISO/IEC 42001: An AI Management System for Industry
The first international standard for managing AI is a management system, not a model test - and for a plant that distinction is the whole point.
Optimizing Aeration in Water and Wastewater Plants
How a sensor-to-model control loop trims the biggest power load in an activated-sludge plant without risking the permit.
Copilots in the Control Room: What the Demo Skips
An AI copilot is a useful reference librarian for operators, not an engineer - here's where it earns its keep and where it has to be fenced out.
OPC UA FX at the Field Level: Past the Gateway
What the OPC Foundation's Field eXchange profile actually standardizes between controllers, the TSN and Ethernet-APL wires underneath, and where it isn't ready yet.
Thermal Imaging for Predictive Maintenance
How infrared inspection turns invisible heat into lead time before a connection fails.
From Historian to Lakehouse: The OT Data Pipeline
How years of plant tag history move out of the historian into an open, queryable lakehouse, walked layer by layer from sensor to served model.
Turning Plant Data Into a Product Carbon Footprint
How to turn historian energy and throughput data into a supplier-specific Scope 3 footprint an auditor will accept.
Catching Scrap Early in Battery Gigafactories
Why cell plants still scrap 15-30% of early output, and how inline measurement and cell genealogy move defect detection from the gate back to the coater.
The Asset Administration Shell Gets a Deadline
Industry 4.0's standardized digital twin was optional for a decade. EU product law just made it a market-access requirement.
MQTT, Sparkplug B, and the Unified Namespace
How one MQTT broker, the Sparkplug state model, and an ISA-95 topic tree replace a plant's point-to-point integration sprawl.
The EU AI Act on the Plant Floor
Who counts as a deployer, when plant AI is high-risk, and the deadlines that land on the factory floor.
Agentic AI on the Plant Floor: What the Demos Skip
Six claims that ride in on the agentic AI procurement deck, and why each one breaks against a deterministic control room.
Industrial Data Governance: Owning OT Data
Who owns the data your machines produce, who may see it, and how to keep it trustworthy from the sensor to the boardroom.
Energy per Tonne: Benchmarking Specific Energy Use
Why specific energy consumption is the metric that actually exposes plant efficiency, and how to benchmark it without lying to yourself.
Cold Chain Monitoring: 5 Myths That Cost Product
Five things vendors and managers get wrong about cold chain temperature monitoring, and what actually catches an excursion in time.
Industrial AI in Food & Beverage: Yield, Uptime, Quality
Where machine vision, soft sensors, and predictive maintenance actually have to run on a regulated production line.
Digital Twins in Process Plants: Where They Pay Off
Telling a synchronized twin from a dashboard, and a four-line test for whether a process plant should build one.
Computer-Vision Scrap Sorting: Cleaner Charge, Less Copper
Cameras grade scrap by sight, but it takes spectroscopy to read the copper that wrecks a heat, and a control loop to act before the grab closes.
NIS2 for Industrial Operators: What EU Plants Must Do
How the EU's NIS2 directive turns into concrete OT controls on a processing plant floor, built in the order that actually works.
Choosing a CEMS: Extractive, In-Situ, or PEMS
Continuous emissions monitoring isn't one technology. Pick extractive, in-situ, or predictive on accuracy, latency, availability, and who maintains it.
Soft Sensors: Keeping an Inferred Measurement Honest
Choosing, validating, and maintaining inferential sensors that survive their first feedstock change.
Securing Industrial Control Systems with IEC 62443
How to cut a processing plant into zones and conduits, set a security level for each, and build a boundary that actually holds.
Combustion Efficiency on a Waste-to-Energy Boiler
Why squeezing the burn on a plant that runs on garbage is a control problem you fight every shift, not a setting you dial in once.
Computer-Vision Quality Inspection: Five Myths
A vision cell sees what a tired inspector misses, but only if the engineering and the claims around it are honest.
Reheat Furnace and Rolling Mill Optimization in Steel
Where the hot end's energy, yield, and quality really trade off, and the instrumentation that holds the line.
CIP Optimization: Cut Water and Time, Keep the Clean
How instrumented return lines turn over-cautious clean-in-place cycles into measured ones, without ever shipping a failed clean.
Anomaly Detection on Process Data: Before the Alarm Trips
Fixed alarm limits catch the gross excursion and miss the early drift. Here's how learned anomaly detection scores raw sensor streams, and why the threshold, not the model, is the hard part.
ISA-95 in Practice: Wiring the Plant Floor to the ERP
How the standard maps the plant-to-business seam, what it actually standardizes, and where real integrations slip.
MTBF and MTTR: Four Myths That Hide Downtime
Reliability metrics are only as honest as the failure definition and the timestamps behind them. Four claims worth dismantling.
Process Historian vs Time-Series Database
Two ways to store plant data, compared by storage cost, query, integration, and fit - and how to tell which one your operation actually needs.
Edge vs Cloud for Industrial Analytics
Stop deciding edge-versus-cloud plant-wide. Decide it per workload, on latency, bandwidth, reliability, and who maintains it.
OEE for Process Operations: What Actually Limits Throughput
Overall equipment effectiveness travels to a continuous plant only if you stop counting parts and start measuring flow, loss, and the constraint.
OPC-UA, Modbus, or MQTT: Picking a Plant Protocol
A working engineer's comparison of the three protocols that run plant networks, scored on the criteria that actually decide a deployment.