Skip to content
Koidra
Book a demo

Process control, energy and maintenance

Manufacturing

Raw materials and operating conditions vary from load to load. Moisture, particle size, composition and temperature can change with the supplier, season and batch, and every machine downstream has to absorb that variation. When the process begins to drift or equipment starts to degrade, the warning often appears first as a gradual trend rather than an alarm. If those trends are not recorded, connected and reviewed in context, problems may only become obvious after energy use rises, product quality shifts or equipment trips.

Book a demo

What goes wrong, and what it costs

Variable inputs move the process away from its target
Changes in moisture, particle size, composition or temperature alter the amount of thermal and mechanical work required. If the process does not adapt, energy use, throughput or product quality can drift.
Blend or recipe changes propagate through the line
A change in material mix can increase the load on one machine and change the conditions seen by the next. Without batch context, the cause may be difficult to trace after performance begins to move.
Condensation develops in the gas path
When gas temperature approaches the dew point, moisture can condense and cause dust or material buildup on downstream equipment. Over time, that buildup can affect airflow, balance, and equipment reliability.
Equipment condition degrades gradually
Vibration, current or temperature may worsen over days or weeks while individual readings remain below alarm limits. Looking at the trend together with maintenance history can reveal degradation earlier.
A sensor stops responding without reporting a fault
A frozen or drifting sensor can continue reporting a plausible value while no longer representing the process accurately. Detecting abnormal sensor behavior is therefore part of maintaining reliable process data.

Proof

Ayo Biomass runs on Koidra across multiple sites, and its dryers gained 20% in operational efficiency after a two-phase KoPilot deployment. The failure modes on this page are the ones their engineers actually write root cause analyzes about, generalized to the sector and stripped of anything specific to one product line.

What process plants have in common

Manufacturing processes differ widely, but several operational challenges appear again and again.

Inputs vary, and the process has to absorb that variation
Moisture, composition, particle size and temperature can change between suppliers, seasons and production batches. A control strategy that assumes constant input conditions will not always produce consistent results.
Efficiency can drift without triggering an alarm
Nothing has to fail for energy per tonne to increase. A gradual change in operating conditions can raise energy use for days or weeks before it becomes obvious in production reports or utility costs.
Equipment condition changes gradually
Vibration, temperature, current and other signals can begin moving long before a machine trips. Looking at those trends together with maintenance history makes gradual degradation easier to identify.
Where a process plant loses groundVariable inputs enter a dryer, mill or boiler that has to hit a specification. Two things drift underneath without an alarm: energy per tonne, and the condition of the equipment.Inputs varymoisture, composition, size,supplier to supplierDryer, mill, boilerhas to absorb the variationProduct to specificationEnergy per tonne driftsfor weeks, with no alarmEquipment condition slipsvibration and temperature creep up
Variation comes in at the front. What it costs shows up underneath, in energy per tonne and in the equipment, long before any alarm.

Where we go deepest

Koidra is particularly suited to energy-intensive process equipment such as dryers, mills, boilers, kilns, furnaces and the thermal utilities around them.

These systems share a common control problem: variable material enters the process, equipment converts it toward a target specification, and operators must continuously balance throughput, quality, energy use and equipment limits.

The feedstock yard of a wood pellet plant, hills of wood chips and sawdust beside the drying hall, with a loader carrying a bucket in.
Wood chips and sawdust waiting for the dryer at a pellet plant. The moisture in this pile is the first number the process has to absorb.

Works with the control system you have

Koidra connects to the PLC and SCADA systems already running your plant and reads the data they already hold. That data, with the notes your operators add beside it, goes in front of the people who decide, so a decision that used to wait for the weekly report is made on what the plant is doing now.

When you are ready for KoPilot, the same connection carries setpoints back to the SCADA, autonomously, inside the limits your engineers set. Everything before that runs on read-only access.

Which systems and protocols we read and write is on the integrations page.

Koidra and the control system you already haveKoidra reads the data your PLC and SCADA already hold and puts it in front of the people who decide. When the plant is ready for KoPilot, setpoints are written back to the same SCADA, autonomously.the data they already holdsetpoints, when you areready for KoPilotYour PLC and SCADAalready running the plantKoidradecisions made on live data,not on last week's report
Nothing is re-instrumented. The plant keeps its PLC and SCADA, and gains the decisions that were waiting on a report.

Failures develop before they become events

Not a sudden event. A shape that repeats for weeks while every individual alarm looks survivable.

Vibration may increase over several days, fall after cleaning or maintenance, and then begin rising again. Each individual alarm may appear manageable, but the repeated pattern can indicate a developing reliability problem.

Seeing that pattern requires more than an isolated alert. It requires enough historical data to show the trend and maintenance records on the same timeline to show what happened between one event and the next. That is the value of bringing process and maintenance data into one operational record.

How a failure announces itself for weeks A vibration trend over several weeks, drawn as a sawtooth. Each cycle climbs as dust builds on the rotor, drops when the machine is cleaned, and starts the next cycle higher than the last. An alarm threshold runs across the top of the chart. Every individual peak sits below it until the final one crosses, at which point the machine fails, even though the trend had been rising for weeks. alarm threshold cleancleancleanclean it breaks week one week six Vibration, cleaned four times, never investigated once
Each clean resets the symptom and none of them touches the cause, so the trend returns higher every time. Four separate work orders, each closed on its own, because the thing connecting them is a line no maintenance system was looking at.

What each product gives a plant like yours

Two ways in, a layer that needs both, and control on top. Each is a complete product with its own payback. Here is what they do here rather than in general.

  1. Start here

    DataPilot

    Mill current, dryer outlet temperature and energy per tonne on one clock, with the moisture readings your shift team types in beside them. A dew point margin nobody was comparing becomes a number you can set an alarm on.

    The DataPilot alert history filtered to one hammermill. Three alerts over four days, one critical and ongoing, one resolved, and one muted by a named responder who wrote down why.

  2. Start here

    Maintenance Management

    Vibration climbing over weeks opens a work order against the fan itself, with its own history attached, instead of four alerts each closed as a separate event. Preventive work is scheduled on running hours and tonnes, not a calendar.

    The Maintenance Management work order board. Sixty-three open jobs on the left, and one opened on the right with its asset, location, work type, source and the procedure attached.

  3. Comes with both

    AI Advisory

    A weekly account of where energy per tonne went, by line and by product, and the follow-up answered from the same record: which supplier, which batch, which shift, and whether a setpoint moved. And what to change to bring it down.

    What one record produces, and what you put in The whole operating record, holding sensors, shift notes, work orders and production batches, produces four things: a daily note, a weekly summary, a recommendation on what to change against the KPIs you set, and a work order opened early when a trend shows a failure coming. Your KPIs and your questions go in, and both are answered from the same record. The whole operating record sensors, shift notes, work orders, production batches Daily note what moved since yesterday Weekly summary and what it cost What to change against the KPIs you set Work order, early a failure seen coming Your KPIs and questions as numbers, or in words arrives without being asked for both answered from the same record
    Two reports run on a timer, a recommendation says what to change against the KPIs you set, and a work order opens when a trend says a failure is coming. Your KPIs and your questions go in, and what comes back is drawn from the same record everything else was built from.

  4. Autonomous control

    KoPilot

    Combustion air ratio and dryer setpoints moved against the moisture actually arriving, inside an envelope your process engineers set, so the line absorbs input variation rather than tripping on it.

    The closed loop, and the envelope it obeys The process is read by a physics-aware model, whose proposed setpoints pass through an operating envelope that states the limits it may not cross, then reach the controllers, which write them back to the process. The loop runs every minute. The process dryer, boiler, hall, crop Physics-aware model knows what the equipment can and cannot do Operating envelope the limits it may not cross Controllers setpoints written back every minute, not every shift
    The envelope is a separate box on purpose. It states what the loop may not do, and it is engineered for the site rather than learned from it, which is why a model that has never seen your plant is still safe to run on it.

What we measure here

See Koidra on your plant data

Bring a week of readings from one line. We will show you what the platform sees, what it would have flagged, and what it would have scheduled.

Book a demo