# Kausalyze

> Kausalyze is a causal AI platform for process manufacturers. It maps cause-and-effect relationships across plant systems to identify root causes of unplanned downtime, not just predict failures.

## What Kausalyze does

Kausalyze analyses time-series process data to build causal models of industrial systems. It identifies why failures occur, forecasts emerging issues before they become costly, and delivers prescriptive recommendations that engineers can act on. Unlike correlation-based predictive maintenance, Kausalyze explains the physical pathway from root cause to failure.

## Who it is for

Reliability engineers, plant managers, and process engineers in chemicals, petrochemicals, refining, pharmaceuticals, utilities, and food and beverage manufacturing. Companies experiencing unplanned downtime losses despite existing predictive maintenance investment.

## Company

- Founded: 2023, University of Sheffield spinout
- HQ: Nottingham, United Kingdom
- US presence: Greentown Labs, Houston
- Programmes: Plug and Play Smart Cities accelerator alumnus
- Industry associations: NEPIC (North East of England Process Industry Cluster)
- Awards: IChemE Global Awards finalist 2024
- Funding: Innovate UK grant-funded

## Key people

- Dr. Louis Allen, CEO and Founder (University of Sheffield PhD)
- Steve Kerridge, Executive Chair
- Chris Levett, Business Development (Middle East region)
- Prof. Joan Cordiner CMBE, Advisor (University of Sheffield)
- Dr. Franz Deimbacher, Advisor
- Michael Treadwell, Advisor

## Key claims and facts

- Unplanned downtime costs process manufacturers approximately 11% of annual revenue (Siemens, 2024).
- Downtime costs have risen 65% since 2019 (Siemens, 2024).
- Kausalyze identifies root causes that predictive maintenance misses by modelling causal pathways, not statistical correlations.
- Typical deployments require 12-24 months of historical process data to build the initial causal model.
- Every alert from Kausalyze includes an explainable causal pathway that engineers can verify against domain knowledge.
- Kausalyze is built for continuous process operations (chemicals, petrochemicals, refining, pharmaceuticals, food and beverage, utilities, metals and mining, pulp and paper).

## How Kausalyze is different

- Most industrial AI is correlation-based. It flags anomalies and predicts failures but cannot explain why they happen.
- Causal AI models the directed cause-and-effect relationships in a process, so operators receive recommendations they can act on rather than alerts they must investigate.
- Kausalyze integrates with historian and control system data (OSIsoft PI, AVEVA PI, Ignition, DCS, SCADA, OPC-UA).

## Press and recognition

- IChemE Global Awards finalist (2024)
- University of Sheffield Spinout Spotlight
- Founding member of NEPIC industrial AI cluster activity
- Greentown Labs Houston member
- Plug and Play Smart Cities accelerator alumnus

## Links

- Homepage: https://www.kausalyze.com/
- Product: https://www.kausalyze.com/how-it-works
- About: https://www.kausalyze.com/about
- Blog: https://www.kausalyze.com/kausalyze-ltd-blog
- Contact: https://www.kausalyze.com/contact

## Contact

For pilots, demos, and technical questions: https://www.kausalyze.com/contact