Step 01
Pretrained once
Learns from many synthetic datasets where the true causes are known, including many with no real relationship.
AI PROCESS SOLUTIONS™ · ROOT™ CAUSAL AI
Root™ finds what is really driving your plant.
Temporal and causal AI that diagnoses problems across the whole plant from the data you already collect: secure, advisory-only, and delivered by process engineers who know your equipment.
Root is the lead technology AI Process Solutions™ (AIPS) supplies to Applied Catalysts + Technologies’ brands. It runs on a temporal causal prior-data fitted network (TCPFN): causal AI that was pretrained once, reads a new plant without being trained on it, and tells engineers how sure it is about every link it finds.
Many teams are working on causal models or on time-series models. TCPFN is built to do both at once, which is what plant data needs: cause and effect, unfolding over time.
WHAT EACH PART OF YOUR TEAM GETS
Why plants choose Root™
Answers that move the P&L
FOR THE CEO, CFO, FINANCE AND CAPITAL PROJECTS
Deep problem diagnosis. Causal chains expose multivariable
impacts, even when the plant no longer behaves the way theory
says it should.
Plant-wide profit, not one pipe at a time. Finds problems and
bottlenecks fast, even complex problems with multiple causes.
Keeps knowledge as people retire. Captures what
experienced operators know before it walks out the door.
Beyond the process. Process-agnostic, including commercial and supply chain data.
Fast to start, easy to live with
FOR IT, ENGINEERING, OPERATIONS, QUALITY AND MAINTENANCE
Fast to set up. Zero-shot: Root reads your plant without building a custom model first. Hosted and self-managed
options.
Easy to use. Works from your historian or simple CSV exports,
plus the documents you already have.
Easy to maintain. Zero-shot capability keeps the model working even after a new unit is added or a repair is made.
Improves over time. Added context and your engineers’ feedback make each answer sharper.
Secure and always in your control
FOR THE CIO AND CEO
Human in the loop, always. Advisory only: no writes and no control actions. Your engineers check every finding against the process physics and safety policies.
Your data stays yours. Nothing is stored on our side, and your data is never used to train our models.
Meets your security requirements. Options to anonymize data,
deployed in your cloud or on premises, SOC 2 compliant, with a
fully air-gapped on-site option for the most secure
environments.
Delivered by process engineers
FROM APPLIED CATALYSTS + TECHNOLOGIES
Efficiency from day one. We help you choose the right data points and objectives before any analysis runs.
Faster changes. Findings connect directly to consulting, equipment, catalyst and services support.
Implementation by people who’ve run plants. Our process engineers turn findings into practical actions your operators and maintenance teams can carry out, not just a data report.
AI is a family of tools
When most people say “AI,” they mean generative AI, the chat tools that write text, code and images. Generative AI is one member of a larger family, and the right tool depends on the problem you are trying to solve.
Generative AI
Creates new content: text, code and images. Our brands use top-tier LLMs every day for research, writing and analysis.
Predictive AI
Learns from past data to forecast what comes next. Credit scores and weather forecasts are everyday examples.
Casual AI
An emerging field that finds cause and
effect: what is driving an outcome, and
what changes if you act on it.
These core types are combined into AI built for specific jobs. Physical AI runs robots and autonomous vehicles. Conversational AI powers chatbots. Industrial AI works with plant historians, SCADA, MES and PLCs to improve processes. Industrial AI is where Root works.
Why generative AI can’t do the job by itself
Wiring a general-purpose chat model to a plant historian is a common first step. Here is how that
compares with Root, which pairs time-series analysis with causal AI.
| Generative AI alone, wired to the plant | Root (time series + causal AI |
| What it reads - Words: manuals, reports, code and data you upload | Plant sensor data, every tag at once |
| What it finds - Likely causes, based on textbook knowledge | The drivers in your plant’s own data: how big, how soon and how sure |
| Cause or coincidence - Can mistake one for the other |
Trained to separate cause from coincidence, with a confidence score on each link |
| A new plant - A new project each time | Pretrained, and works from day one |
| Security - Top-tier models typically run in the provider’s cloud, with plant data sent on every call. Strip out the tag names and the model loses what it reasons from |
Can run on premises, inside the plant’s own network, or work from anonymized data |
| Cost to run - Charged per use, and grows with use | Fixed, on one small model |
| Guardrails - You build approvals, audit and testing | Built in: approvals, audit and verification |
| Time to value - Six months or more to design, build, test and maintain | Weeks: pre-built, pretrained and simple to adapt |
One situation, three answers
Picture an acid plant where SO₂ at the stack keeps creeping up, and no one knows why.
GENERATIVE AI CHATBOT
“Why is my SO₂ rising?”
Answers with the likely causes from textbook knowledge.
PREDICTIVE AI FORECAST
“What happens next?”
Warns that the plant will likely exceed its permit limit again this month. Useful, but it comes without a fix.
PREDICTIVE AI FORECAST
“Why, and what should I change?”
In the plant’s historian data, SO₂ rises this way after an 8 °C drop in bed 1 inlet temperature that follows gas-strength changes. The fix: raise the inlet setpoint with each change.
Illustrative example, not results from a specific plant.
THE TECHNOLOGY BEHIND ROOT™THE TECHNOLOGY BEHIND ROOT™
Why TCPFN is the right technology
Most industrial AI falls into one of three groups. Large language models learn from text. Whole-plant models learn from years of one plant’s history so they can predict and optimize it. First-principles models are built by engineers from the equations for the variables they choose to include. Each has its place. None was designed around the question plants ask most often: what caused this?
Many of these models are also custom builds. Each one takes months of data collection and cleanup, fits one unit at one site, and has to be built again for the next. Foundation models changed that pattern in other fields: one pretrained model works on new problems without being rebuilt. Plants needed the same thing for cause and effect. Earlier causal foundation models worked only on static tables of data, with no way to handle readings that unfold over time [1]. Almost all plant data unfolds over time.
Root runs on a temporal causal prior-data fitted network (TCPFN), an architecture published in June 2026 by our technology partner, ProfitOps [1]. Root then links the relationships it finds into ranked, multi-step chains of events, building on the same team’s earlier published work on root cause analysis [2].
01 Trained once, then used as-is.
Pretrained on a very large number of synthetic datasets where cause and effect were known. It answers in a single pass, with no training, fitting or tuning on your plant.
02 Trained to say “no effect."
Much of its training data contains no real relationship at all, which teaches it to reject noise and shared drivers instead of reporting every correlation.
03 It reports how sure it is.
For every candidate link it estimates whether the effect is real, whether a hidden common cause may be behind it, whether the data can settle the question, and whether the link is direct, indirect or part of a feedback loop.
04 It returns a short list.
On one published benchmark with 104 true relationships, it returned 160 candidates. A standard Granger causality method returned 6,164 [1].
05 It keeps plant data private.
TCPFN works from the numbers alone, so tag names and units can be stripped out before analysis and matched back afterward on the plant’s side.
How TCPFN finds cause and effect
Step 02
Reads the plant in one pass
Sensor and historian data goes in. No training, fitting or tuning on that plant.
Step 03
Scores every link
Is the effect real, is a hidden cause behind it, can the data settle it, and what kind of link is it?
Step 04
Builds the causal map
Root chains the strongest links into ranked, multi-step chains of events [2].
Step 05
Engineers check the physics
AC+T process engineers test each link against the process physics before anyone changes the plant.
TCPFN by the numbers
Published results from the TCPFN research paper [1].
19
Benchmark datasets across five domains, all run with the same model and no tuning
160
Candidate links on a benchmark with 104 true relationships, against 6,164 from a standard Granger method
259 MB
Pretrained model, small enough to run inside a plant’s own network
~6 hrs
To analyze 1,275 variables from an industrial plant on a single GPU
In the field: Root traced recurring sheet breaks at a U.S. pulp and paper mill to an upstream cause the mill’s experts confirmed, with about three hours of median early warning.
Why it fits process plants
Chemical and materials plants change all the time. Catalysts age, exchangers foul, feedstocks shift, campaigns switch and units get debottlenecked. A model trained on a plant’s past starts to go stale when the plant changes. TCPFN is never trained on the plant, so a change simply becomes part of the data. Move the time window and run it again.
Well-run plants hide their own cause and effect. Good control removes the variation a model needs. TCPFN says when the data cannot settle a question, so engineers know when a short, targeted plant test is the better next step.
Plant data is sensitive. Root can run on premises, with read-only access and no outbound connection, or work from anonymized data, so the plant decides what leaves the site.
Engineering time is limited. A short, ranked list with a confidence score on every link fits inside an engineer’s week.
Findings need someone to act on them. AC+T’s brands design catalysts and materials, engineer processes and build equipment. When Root points to a cause, the same teams can help fix it.
Your expertise stays yours
Every plant runs on know-how that lives in people’s heads. When experienced operators retire, much of it leaves with them. Deloitte and The Manufacturing Institute estimate that U.S. manufacturing could need as many as 3.8 million new employees between 2024 and 2033, and that as many as 1.9 million of those jobs could go unfilled [3].
There is a second, quieter leak. When experts correct a general-purpose AI tool day after day, each correction carries real judgment. With many hosted tools, that learning stays with the provider, and it leaves when the contract does.
The model stays the same. The pretrained model does not learn from plant data, and that data is never used to train it.
The plant’s knowledge grows. Causal maps, confirmed causes, key operating parameters, and the context and feedback engineers add all build up for that plant.
A question worth asking any AI provider: if you leave, what happens to everything the system learned from your people?
HOW IT WORKS
01
Scope
Pick the problem and agree the objective with your engineers.
02
Share data
Read-only historian or CSV exports, plus event logs and drawings.
03
Diagnose
Root™ maps cause and effect and ranks the likely drivers.
04
Validate and act
Engineers confirm against the physics; you decide what to change.
WAY TO WORK WITH US
Simple Projects
A focused study on one problem, with a defined scope and report.
Root™ subscription
Ongoing causal monitoring and diagnosis, with predictive add-ons.
Special projects
Custom engagements for complex or plant-wide questions.
INDUSTRIES WE SERVE
Sulfuric acid
Specialty chemicals
Toll manufacturing
Pharmaceuticals
Specialty materials
Oleochemicals
Government
Why AC+T
Physics and data on one team
Through ActiveSite, our engineers build process models, kinetic models and DFT calculations, so every finding is checked against the physics.
Process engineers behind the technology
Our people have run plants. They know which problems are worth solving and how to turn a finding into action.
Findings connect to fixes
Our brands supply catalysts and materials, process engineering and custom equipment, so a confirmed root cause has a direct path to a solution.
Secure choices
Root is SOC 2 compliant and can run on premises or on anonymized data.
FAQ
What is TCPFN?
TCPFN stands for temporal causal prior-data fitted network. It is a causal AI model pretrained once on synthetic cause-and-effect data, built for time-series data like a plant historian, and published in June 2026 [1].
Does Root need to be trained on our plant’s data?
No. TCPFN reads a new plant in a single pass, with no plant-specific training, tuning or retraining. When the plant changes, you move the time window and run it again.
How is causal AI different from generative or predictive AI?
Generative AI explains likely causes from general knowledge, and predictive AI warns you something is coming. Causal AI finds what is driving the outcome in your own plant’s data, and how sure it is about each link.
Does Root ever change our plant?
No. Root is advisory only: no writes and no control actions. Your engineers review every finding and decide what to change.
How does Root keep plant data secure?
Root is SOC 2 compliant and can run in your cloud, on premises inside the plant’s own network, fully air-gapped, or on anonymized data. Tag names and units can be stripped out before analysis and matched back on the plant’s side.
Why does a small model matter?
The pretrained TCPFN model is 259 MB and ran a 1,275-variable industrial dataset in about six hours on a single GPU [1]. A model that size can run inside a plant’s own network, at a fixed cost.
Start with one problem you already argue about.
Bring a recurring upset, bottleneck or disputed change. We scope it with your engineers, run Root™ on your existing data, and walk you through what is really driving it.
Investors and technology partners: we’re glad to go deeper on TCPFN, how Root is built and where we’re taking it.