DATA ANALYTICS THEME
Drupad Trivedi; Lucy Morgan
-
Embedding AI-Driven Predictive Analytics for Sustainable Process Optimisation
Rationale: Analytical chemistry generates vast, heterogeneous datasets from techniques like LC-MS and NMR, yet the UK industry lags in leveraging AI to predict environmental impacts and optimise resource use in real time. This challenge addresses the push for net-zero goals under the UK's Environment Act 2021, where data analytics can reduce waste in chemical manufacturing by forecasting degradation pathways or solvent efficiencies, aligning with CAMS's focus on innovation disconnects.
Defining the Grand Challenge
This grand challenge focuses on embedding AI-driven predictive analytics seamlessly into UK analytical chemistry workflows to achieve sustainable process optimisation. This means not just accurate predictions, but fully integrated, real-time adaptive control that optimises resource use, minimises waste, and aligns with green chemistry goals—across the diversity of UK chemical industries and analytical setups.
Current State of the Art and Limitations
While leading AI methods (deep learning, hybrid chemometric-AI models, digital twins, real-time monitoring, and explainable AI) have demonstrated promising results in parts of chemical process analytics, they largely remain in research pilot phases or applied in isolated contexts. Existing limitations include integration gaps, generalisation issues, lack of sustainability focus and lack of human-AI interactions.
AI models are often not fully embedded within live, industrial analytical workflows due to challenges in data interoperability, process heterogeneity, and legacy equipment connectivity. Many AI models lack robustness when transferred across different labs, instruments, or conditions, limiting wide-scale deployment across UK industry. Quantitative incorporation of sustainability metrics (energy, waste, carbon footprint) into predictive analytics remains underdeveloped and rarely influences real-time control strategies. Trust and usability of AI tools by chemists and engineers are often insufficiently addressed, slowing adoption and integration into decision-making processes. Thus, the gap lies in creating a holistic, robust, and sustainable AI-driven predictive analytics framework that is fully integrated and interoperable with diverse analytical chemistry instrumentation, end users and data streams.
The Grand Solution
The envisioned solution is a modular AI ecosystem that combines multi-source data ingestion, adaptive predictive modelling, digital twin simulations, and sustainability-focused optimisation algorithms. This ecosystem would provide process chemists and engineers with intuitive, real-time decision support platforms that predict process deviations, recommend corrective actions, and quantify environmental impacts dynamically. Unlike current fragmented tools, this integrated approach would facilitate widespread industrial uptake in the UK, driving measurable sustainability improvements company wide. We welcome targeted proposals addressing this the grand challenge by synthesis and scaling of state-of-the-art capabilities into a deployable, sustainable AI system tailored for UK analytical chemistry, responding directly to the gaps identified in current technologies. It aligns closely with CAMS UK's mission in data analytics and sustainability.
Aims
- Establish an integrated AI-driven predictive analytics framework that operates within UK analytical chemistry workflows to enable real-time sustainable process optimisation across diverse instruments and sites.
- Demonstrate measurable reductions in waste and energy use while maintaining or improving analytical performance, aligned with net-zero and Environment Act 2021 ambitions.
- Build capacity within CAMS UK for data-driven innovation, industry–academic collaboration, and scalable adoption of AI-enabled analytics.
Objectives
- Map heterogeneous data streams from techniques such as LC–MS, NMR, and related instruments; assess data quality, provenance, and interoperability requirements.
- Create models combining machine learning with chemometrics designed for transferability across laboratories and equipment.
- Explicitly couple sustainability metrics (energy consumption, solvent use, waste generation, carbon footprint) with model objectives and decision-support outputs.
- Investigate digital twin concepts and adaptive feedback loops to inform decisions that optimise sustainability without compromising data integrity.
- Develop intuitive interfaces and governance frameworks to foster trust, explainability, and safe deployment across CAMS networks.
- Plan cross-site pilot studies to test generalisation, interoperability, and sustainability gains in representative UK settings.
COMPLEX MIXTURES – SEPARATION & DETECTION THEME
Mansoor Saeed; Ali Salehi-Reyhani
1. Integrated Separation Science Solutions for Complex Analytical Challenges
The core challenge for separation science lies in effectively handling complex mixtures and achieving precise detection across diverse molecular systems.
The challenge focuses on:
- Novel Separation Processes: Moving beyond traditional chromatography, developing efficient, alternative techniques for complex mixtures such as biotherapeutics, agrichemicals, and more.
- Flexible & Adaptable Systems: Aim to create multi-dimensional and modular separation platforms utilizing innovative materials and technologies, capable of handling diverse molecular scales and properties.
- Advanced Detection & Characterization: Iimplementing state-of-the-art, high-sensitivity analytical detection technologies. Iintegrating AI and machine learning for data processing and characterization, which minimizes reliance on traditional reference standards.
- Intelligent Separations (Predictive Modelling & Simulation): AI integration for enhanced predictions and computational models, optimizing separation outcomes and accelerating method development.
The challenge is to enhance and broaden separation science, enabling the quantification of molecular systems across different length scales and complexities. The goal is to establish analytical platforms that meet modern analytical separation requirements, extending the scope of separation science to include biomolecules and coordinated chemical machines, thereby providing comprehensive, cutting-edge, and adaptable solutions.
2. Mastering Complex Mixtures with Integrated analytical separation science system
The paramount analytical challenge in modern separation science is the development of integrated, intelligent analytical systems capable of efficiently separating and precisely detecting components within highly complex mixtures, especially those involving diverse molecular species and requiring minimal reliance on traditional reference standards.
Example: Characterise a Novel Biotherapeutic Drug
- The Problem: Traditional separation methods often struggle to resolve all these critical micro-heterogeneities. Furthermore, accurately detecting and quantifying these low-abundance variants, particularly when specific reference standards for each variant are unavailable, is a significant hurdle.
- An Integrated Solution: An integrated analytical system that combines multi-dimensional separation platforms (e.g., coupling different chromatographic modes for enhanced resolution) with advanced, high-sensitivity detection technologies like high-resolution mass spectrometry. AI and machine learning are integrated to process the vast data generated, identify and characterize unknown variants, predict their behavior, and even optimize separation conditions in real-time. This approach minimizes the reliance on traditional reference standards and provides a comprehensive, deep characterization of the biotherapeutic's heterogeneity, ensuring both safety and efficacy.
The above example exemplifies how broadening of separation science, enabling precise quantification of molecular systems across different length scales and complexities through cutting-edge, adaptable, and intelligent integrated separation science solution.
SENSORS AND PHOTONICS THEME
Binoy Paulose; Natalie Belsey
1) Development of next generation sensing modalities for point-of-use applications
Aim:
This challenge aims to advance the fundamental scientific understanding necessary to create new sensing modalities capable of enabling precise, real-time detection at the point of use, by uncovering novel physical, chemical, and biological transduction mechanisms and defining the principles that govern their performance.
The goal of the challenge is to aid the development of next generation point-of-use sensing platforms to improve the quality of life.
Objectives:
- Development of novel sensing mechanisms by identifying unexplored fundamental physical, chemical, or biological interactions that can be leveraged for point-of-use sensing.
- Development of theoretical models that describe signal generation, propagation, and detection within these new modalities, enabling predictive understanding of sensing behaviour.
- Identification of materials, interfaces and data interpretation tools to enable the adaptation of these new sensing modalities for point-of-use applications
2) Photonics based solutions for in-line/in-vivo monitoring of processes
Aim:
This challenge focusses on developing photonics-based solutions for in-line and in-vivo monitoring of processes across various fields including manufacturing and healthcare.
In manufacturing, these solutions can be aimed at enabling non-destructive and real-time monitoring, facilitating precise control and optimization of processes by continuous monitoring of process parameters to ensure product quality and safety while minimizing waste and production downtime.
In healthcare, this challenge can be aimed at developing tools for non-invasive in-vivo monitoring of biological processes within tissues and organs in real-time,
Objectives:
- Development of photonics-based solutions for accurate and reliable measurements in various environments, from biological tissues or to industrial manufacturing sites.
- Development of compact, high-performance photonic devices suitable for in-vivo applications, which requires advancements in miniaturization while maintaining measurement capability, integration with existing systems, and overcoming limitations in size, weight, and power consumption.
- Development of affordable and scalable photonic devices for widespread adoption in both medical and industrial applications.
