We build Physical AI for the process industries. DecaOS, our proprietary “process reasoning model”, works with self-driving labs that design and run their own experiments, and with running plants, to develop cleaner processes and keep every plant at its best. Energy use and CO₂ are design targets from the first run, not checked afterwards.

No. Your engineers stay in charge: they set the targets and limits and approve what runs. DecaOS and the self-driving lab take on the repetitive experiments and the continuous tuning, so people can spend their time on the decisions that matter.

Process historians, control systems (DCS, PLC and SCADA through OPC UA), LIMS, batch records and MES, and process simulators with kinetic and thermodynamic models. A first study can start from CSV or Excel exports, before anything connects. Every connection is listed on our Integrations page.

In process engineering the ground truth is physical. Lab analytics, pilot runs and plant KPIs check every prediction, and DecaOS learns from what actually works. On a live plant, recommendations stay within the limits your engineers set.

No. Most projects start with a study on the data you already have, such as historical exports from your plant or lab, so you can see what DecaOS finds before anything is connected. Self-driving lab campaigns and live plant connections can follow.

You decide what DecaOS can see: only the sources and tags you select, through accounts your team creates and controls. Connections start read-only, and personal data such as operator names can be left out at the source. What DecaOS may learn from your data, where it is stored and how long it is kept are set out in your agreement with us.

Yes, at any time. Revoke the account or disable the gateway connection and DecaOS stops reading your data; we delete the stored credentials. Writing back to the control system is a separate step that your engineers enable for named loops only. See Integrations.

It sets analytics cookies only if you allow them, and you can change your choice at any time in Consent Settings. Messages sent through the contact form go to contact@decaplants.com. See our Privacy notice and Imprint.

At decaplants.si. If you would rather see it on your own process, book a demo and we will walk you through it.

Start by telling us what you want to develop or improve, through the partnership form. Together we set the targets (yield, quality, energy and CO₂) and look at the data you already have. Our self-driving lab then designs and runs the experiments around the clock, and you get the results, the data and a process that is ready to scale. Makers of lab instruments, analytics and pilot equipment can also join the network of instruments our labs run on.

We work in two ways. As a contract research and manufacturing partner, we develop, scale up and produce your process in our self-driving labs, priced per project. Or you license DecaOS for your own R&D teams and plants. The price depends on the scope, from a first study on your data to DecaOS running across your sites. Tell us about your process and we will come back with a proposal, through the contact page.

Pharmaceuticals, food and beverages, pulp and paper, speciality chemicals, petrochemicals, textiles, and energy and environment. You can explore each one in our 3D Industries view.

It is an illustrative scene. The lab, the robots and the plants show how the DecaOS loop works; the physics views are simplified models, and the readings are simulated.

For investor enquiries, write to us at contact@decaplants.com.