Built in pulp & paper. Proven by the names you know.

Q: Which paper manufacturers use ProcessMiner?

Over 20 major paper manufacturers run the platform, including International Paper, Georgia-Pacific, Smurfit Westrock, Metsä Group, Kruger, and Graphic Packaging International. One organization runs about 20 machines in closed-loop control after a trust ramp of roughly 2 years.

Q: What does the lab lag cost a mill?

Lab results on a paper machine lag production by 30 to 60 minutes per roll, so a roll is often underway before anyone knows the last one drifted. By industry benchmarks, a 0.1% quality gain is worth $400,000 to $500,000 a year at an average mill, which is why closing that gap is where most deployments start.

Over 20 major paper manufacturers run ProcessMiner today. This is our home vertical: every model, every outcome metric, and every deployment method reflects a decade of learning on paper machines.

Start a Free Assessment

Trusted on the floor at leading manufacturers

Watch the ProcessMiner pulp & paper overview0:30 · sound on

Read the transcript

Your process data is already telling you what’s wrong. The problem is hearing it in time. ProcessMiner sits on the historian you already run and shows where quality is heading, for the grade you’re making. It ranks what’s driving a deviation, and recommends the setpoint to move to. In open or closed loop, your operators stay in control. One platform, on your own data.

What a paper machine is up against

You lose money on a reel before you know if it’s good

Overshooting spec to stay safe gives away fiber, your biggest cost line (~50% of production cost, industry benchmark). Between turn-ups you fly blind while the lab runs 30 to 60 minutes behind. Shift-to-shift variation means a tough grade runs well only when your best crew is on. Startups and grade changes are where off-spec piles up fastest.

30-second clips

See it on a paper machine

Three short clips, sound on. Each shows one part of the platform at work on paper machine data.

Real-Time Prediction0:30 · sound on

Shows where reel quality is heading every 30 seconds, ahead of the lab.

Read the transcript

Your lab tells you what already happened. By the time the result comes back, the product’s already made. Real-Time Prediction forecasts where quality is heading every thirty seconds, for the grade you’re running. When the trend heads out of bounds, your team gets one specific correction. Inside the band, it stays quiet. So you correct course before product goes out of spec.

PM Vector0:30 · sound on

Shows operators the setpoint back to this grade’s best run, straight through grade changes.

Read the transcript

An alert tells you something’s wrong. Not what to do. So every shift estimates the correction from memory. PM Vector puts your best run for this grade on screen as a live target. When a variable drifts, operators see the exact setpoint to move to, on levers they control. The target retrains as you run, so your best crew’s settings become every crew’s standard.

Root Cause Analysis0:30 · sound on

Narrows 2,000 to 4,000 machine tags to the top 10 likely causes, ranked by influence.

Read the transcript

Quality’s out of spec. Somewhere in hundreds of tags, one or two variables did it. Finding them by hand takes hours. Root Cause Analysis looks at every tag at once, drops the ones along for the ride, and ranks the top ten. One click opens the trend. PM Vector shows the move back. Your team spends the shift fixing it, not hunting for it.

Land and expand: a client story

“A North American paper and packaging manufacturer started with one machine in open loop, operators verifying every recommendation. It expanded machine by machine, mill by mill. About 2 years later, roughly 20 machines run in closed-loop control.“

Trust earned one proof point at a time.

25%

Less wet-strength chemistry

One paper machine, 12 months

~20

Machines in closed-loop control

One paper and packaging organization, after about 2 years

~$1M

Cumulative savings

One paper machine, 12 months

Pulp and paper results from ProcessMiner deployments, each measured against that plant’s own baseline. Results vary; past performance is not a guarantee of future results.

Free Data Health Assessment

Start with one paper machine.

Start with a free Data Health Assessment. We’ll take a closer look and let you know exactly what your own data can do.

Start a Free Assessment

Needs 6–12 months of process history with timestamped lab results.

Questions, answered

Frequently asked questions

Which paper manufacturers use ProcessMiner?

Over 20 major paper manufacturers run the platform, including International Paper, Georgia-Pacific, Smurfit Westrock, Metsä Group, Kruger, and Graphic Packaging International. One organization runs about 20 machines in closed-loop control after a trust ramp of roughly 2 years.

Can one mill deploy without a corporate program?

Yes. Mills operate semi-independently, and most deployments start at the mill level with the mill manager and process engineers, on one machine, not as an enterprise IT program. Expansion to other machines and mills follows the results.

We already run a QCS with centerlining. What does this add?

A QCS measures and controls the sheet against a setpoint, and its centerlining publishes a historical best-run reference. ProcessMiner adds a grade-adaptive best run derived from your own data that keeps retraining, a ranked cause when quality moves, and a quantified adjustment operators can apply. It reads the same PI historian your QCS already feeds, and we are not the QCS vendor, the OEM, or the chemistry supplier.

What does the lab lag cost a mill?

Lab results on a paper machine lag production by 30 to 60 minutes per roll, so a roll is often underway before anyone knows the last one drifted. By industry benchmarks, a 0.1% quality gain is worth $400,000 to $500,000 a year at an average mill, which is why closing that gap is where most deployments start.

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.