Engineering-Grade Explainable AI for Industrial Process Optimisation
About EcoShift.ai
EcoShift.ai was born out of pioneering research at the University of Surrey, with a focus on helping industrial teams understand and improve real plant performance.
We build explainable AI for industrial processes, turning complex operational data into clear insight engineers can trust and act on, especially during process optimisation, plant scale-up and troubleshooting.
The Problem
Many first-of-a-kind industrial processes, from sustainable fuels to cosmetics and food and beverage, fail to perform as expected. Real conditions rarely match design assumptions, leading to instability, delays, and lost performance.
When problems arise, teams rely on slow traditional physics-based models or black-box AI tools that engineers do not fully trust. AI can recommend the best operating parameters, but without clear reasoning, engineers struggle to justify why those settings should be used.
£20-£50M Lost Annually
Due to suboptimal process operation and missed optimisation opportunities.
93% Cite Trust as the Barrier
Organisations say human trust, not technology, is the biggest obstacle to AI adoption.
£2M Lost Per Hour
Caused by delayed troubleshooting decisions.
Our Technology
We combine high-fidelity process models with engineering-grade explainable AI to deliver trusted recommendations for optimisation, troubleshooting, and scale-up, showing not just what to change, but why it works.
Key capabilities include:
Trustworthy, explainable AI decisions
Unlock £30-80M annual value
Ready for engineering sign-off
Scalable across plants and sites
The Team
Dr Xin Yee Tai
Entrepreneur Lead
Xin Yee holds a PhD in Chemical Engineering from Loughborough University and specialises in applying AI to chemical process simulation. Continuing her research at the University of Surrey, she focuses on improving model transparency and interpretability in AI-driven process systems. She leads customer discovery, value proposition testing, and commercialisation strategy, ensuring EcoShift.ai meets industrial needs while driving business development and partnerships.
Professor Jin Xuan
Principal Scientist Advisor
Professor Jin Xuan is a leading expert in sustainable chemical engineering and digital innovation. He is Associate Dean (Research and Innovation) at the University of Surrey and CTO of R3V Tech. As Principal Scientific Advisor to EcoShift.ai, he shapes the technical roadmap, embeds sustainability into process design, and ensures the AI-driven simulation tool meets industrial needs through rigorous validation and collaboration.
John Liley
Business Advisor
John Liley is a Chartered Engineer and senior board advisor with over 40 years of experience in technology commercialisation, strategic planning, and business growth across high-technology engineering sectors. As Business Advisor to EcoShift.ai, he draws on his expertise in IP management, market entry, and funding strategy to mentor the team and strengthen its commercial readiness. John’s extensive track record in guiding start-ups and scale-ups, securing international partnerships, and supporting successful exits brings invaluable insight to shaping EcoShift.ai’s business strategy and long-term growth.
Ross Manning
Technology Transfer Office
Ross Manning is Technology Transfer Manager at the University of Surrey’s Faculty of Engineering & Physical Sciences, with over a decade of experience in technology transfer, commercial strategy, and IP management. He has been instrumental in shaping the commercial roadmap for EcoShift.ai, strengthening the business case and market positioning through feedback on research outputs and presentations. Ross contributes expertise in market analysis, technology scouting, and IP strategy, ensuring the project is well-prepared to engage industry stakeholders and explore viable routes to market.