Quality data collection & machine learning

Enabling

More Efficient

Chemical Processes

What we do

At SOLVE, we specialize in improving chemical processes, driving both efficiency and sustainability. Through our innovative data collection and machine learning approaches, we deliver flexible future proof solutions for superior results.

Unlock the Full Potential of Your Process

Tailored to Your Process, Powered by Real Data

More than just optimization—our flexible solutions are backed by real-world data, supporting you to design the best process for now and the future.

Less expensive, low-risk process development.

The data and models we provide for a customer inform which equipment should be purchased, reducing uncertainty and risk in the decision making framework they utilise.

Implementation of more cost efficient and sustainable processes

Our methods enable more efficient and sustainable processes conditions to be utilised by customers, reducing process operating costs and increasing sustainability.

Shorter time-to-market for products

Our data collection is efficient and effective, allowing customers to reduce process development time cycles and produce a scalable efficient process in fewer iterations.

Evidence to support processes to regulators

Our real-world data and models help justify condition choices to regulators easing regulatory burden and demonstrating the trade-offs between different condition choices.

Backed by

Overcoming Industry Challenges

Why is Finding the Right Process Conditions that Hard?

Data Availability

No existing large process conditions dataset.

There is a lack of comprehensive datasets from which accurate predictions can be made for process conditions. This scarcity of real-world data prohibits companies from taking advantage of the benefits machine learning has brought to other areas of science.

Experimental collection

Data is hard to collect.

Current methods utilised for data collection are inefficient for understanding the effects of process conditions on outcomes, especially for solvents. Building understanding to develop a process is often slow and expensive.

Modeling Accuracy

Theoretical modeling is complex.

Computational models based on theoretical simulation of a process can be useful but are limited in application and costly. Such models can be inconsistent when compared to experimental data, especially when modelling solvent effects.

In the news

Society of Chemical Industry

From PhD to CEO: Data driven chemistry with SOLVE

Chemical & Engineering News

SOLVE uses new chemical process techniques and machine learning to create efficient processes

UK Research and Innovation

Spin-out SOLVE to digitally transform chemical manufacturing

Imperial News

Machine learning to support chemical R&D recognised with best paper award

About Solve

At SOLVE, we are driven by a shared belief that smarter, data-driven solutions can transform the chemical industry and make a real difference in people's lives. Our mission goes beyond simply optimizing processes: we are committed to empowering our clients with flexible, evidence-based options that allow them to apply their expertise in ways that have the greatest impact.

Every member of our team is dedicated to the idea that by improving efficiency and reducing risk, we can help lower the cost of vital products, from pharmaceuticals to agrochemicals, all while driving innovation and setting a new standard for sustainability.

Our

Company

Founders & Directorship

Dr Linden Schrecker
Founder & CEO
Jose Pablo Folch
CO-Founder & CSO
Alexander Giles
Director

Team

Sarah Boyall
Technical Lead
Thomas Dixon
Technical Lead

Advisory Board

Prof. Mimi Hii
Advisor, Professor of Catalysis
Prof. Klaus Hellgardt
Advisor, Professor of Chemical Engineering
Dr. Christian Holtze
Advisor, Academic Partnership Developer
Dr. Joachim Dickhaut
advisor, Senior Principal Scientist

Publications

Automated Optimization of a Multistep, Multiphase Continuous Flow Process for Pharmaceutical Synthesis

10/2024

Operator-free HPLC automated method development guided by Bayesian optimization

06/2024

A comparative study of transient flow rate steps and ramps for the efficient collection of kinetic data

02/2024

An efficient multiparameter method for the collection of chemical reaction data via ‘one-pot’ transient flow

09/2023

Combining multi-fidelity modelling and asynchronous batch Bayesian optimization

02/2023

SnAKe: Bayesian optimization with pathwise exploration

12/2022

Discovery of unexpectedly complex reaction pathways for the Knorr pyrazole synthesis via transient flow

09/2022

Elevate Your Business with

Advanced Chemical Process Solutions

Get in touch