Open loop
Operators apply the platform’s setpoint recommendations, building a track record you can verify.
Autonomous optimization, ready when you are.
Constant manual correction isn’t sustainable, and blind trust isn’t realistic. So every deployment starts in open loop, with recommendations your operators verify and act on, and autonomy grows only in steps you approve.
The adoption path
4 stages. You control every transition.
Every deployment starts in open loop. Each stage opens only when your team is confident, on your timeline.
In open loop, your operator makes the change. In closed loop, the platform writes the setpoint, only on variables you’ve authorized, and the off switch is always yours.
Hourly sprints
Short autonomous runs with an operator on watch, typically 1–2 hours a day during stable production.
Extended runs
Sprints across full shifts, with room for manual override during transitions.
Closed loop
The optional last step: continuous autonomous control that operators can disable instantly.
At each step, a value capture report measures quality, throughput and waste gains against your baseline, so every decision rests on your own numbers.
What you control, always
Autonomy has an off switch. Literally.
You decide when the platform acts, by variable and by moment, and your team can switch it off directly.
Variable-level control
- You choose which variables are eligible, for example chemistry dosing, and which stay manual.
- Variables join closed-loop scope only when your team says so.
- Every autonomous write is logged against the recommendation that triggered it.
Moment-level control
- Toggle autonomy off during a grade transition (Grade A to A1 to A2 to B), then re-enable it once stable.
- Autonomy switches off instantly from the floor, for any reason.
- Closed-loop writes reach your DCS through IoT-agnostic Modbus TCP or OPC UA, or via historian scripts or API.
Autonomous control uses the same PM Vector guidance and quality predictions your team already sees in Real-Time Prediction.
Case study Pulp & paper mill
A client proof case
2 years. One organization. 20 machines.
One client started with recommendations operators could verify, before any setpoint was written automatically. Over roughly 2 years, that track record scaled to about 20 machines in autonomous control.
This is one organization’s path. Autonomy decisions are made at the facility level, and timelines vary with process complexity and history.
Proof point One deployment, chemistry usage
~1,600 lbs
Chemistry saved in one week
At only 32% autonomous usage
Single-deployment results, measured against each plant’s own baseline. Results vary. How we measure →
Free Data Health Assessment
Prove it on your data.
Start with a free Data Health Assessment. We’ll take a closer look and let you know exactly what your own data can do.
Autonomy is optional and gradual. You decide if and when to move past open loop.
Questions, answered
Is it safe, and how much control do we keep?
The questions plants ask before any setpoint is written automatically, answered plainly.
Can ProcessMiner actually control our equipment autonomously?
Yes. Deployments start in open loop, with advisory dashboards and PM Vector setpoints, then move through hourly sprints and extended runs to closed loop when you choose. You decide which variables are automated and which stay manual.
Can an operator override or stop autonomous control?
Yes, instantly, from the floor, for any reason, and without a ticket. Autonomy is a setting your team operates directly: off for a grade transition, a startup, a maintenance window, or because someone wants it off. Open-loop recommendations keep running in the background, so they are still there when you switch back.
What keeps an autonomous setpoint write inside safe limits?
3 things you set. Only the variables you have cleared for closed-loop can be written at all; each write stays inside the limits you define for that variable and grade; and writes go through your existing control architecture, over Modbus TCP or OPC UA to your DCS, historian scripts, or API write-back, so your DCS’s own interlocks, clamps, and alarm logic apply exactly as they do to a manual change. Every autonomous write is logged against the recommendation that triggered it.
How much of the process do we hand over?
As little as you want, for as long as you want. Most plants start hourly sprints with one lever, often chemistry dosing, for 1 or 2 hours a day during stable production, and expand only when the track record supports it. Which variables, which hours, and which grades are your decisions, and each one can be reversed.
What happens during grade transitions?
Operators toggle autonomy off during a grade transition and re-enable it once the process stabilizes on the new grade.
How long does it take to reach closed loop?
It varies by facility. One client reached roughly 20 machines in autonomous control over about 2 years, one proof point at a time. There is no fixed enterprise-wide timeline.
Get in touch
Have a question?
Whether it's our platform, your process, or a free Data Health Assessment you'd like to start, let us know.
Rather talk? Call +1 (470) 910-0770.