Modelling and AI-assisted Safe and Sustainable by Design Product Innovation

Species Distribution Modelling supporting Safe and Sustainable by Design Environmental Assessments 

Leo Posthuma (RIVM)

The European Green Deal’s Chemicals’ Strategy for Sustainability introduced an innovative principle: Safe and Sustainable by Design. This is a pro-active approach, whereby the safety and sustainability of novel chemicals is ascertained in the early stages of development of novel molecules for novel functions. This implies, however, that assessments of exposure, hazard and sustainability metrics need be made for data-poor conditions. There are opportunities to bridge the data gap via Machine Learning/Artificial Intelligence methods. In this presentation, a novel way of using available hazard data will be considered, namely: the comprehensive analysis of a large number of such (species*chemical)-pairs data, via pairwise learning. The assessment method and outcomes were trialled with LC50-data, and yielded a large number of Species Sensitivity Distributions (SSDs), as model often used in Hazard Assessment. SSDs based on those pairwise-learning data are now based on more than 1000 species each. The results can be formatted as innovative Dashboard, where the user enters information on the identity of the studied chemical, to obtain hazard insights on a wide array of species in return. The presentation considers opportunities as well as critical aspects of the approach. Further development (focusing e.g. on chronic test endpoints and much more) is needed to yield optimal outcomes for SSbD-assessments.

AI-assisted Data Collection, Curation and Machine Learning supporting use of Values and Uncertainties in First Tier Safe and Sustainable by Design Assessments

Barry Hardy (Edelweiss Connect)

We discuss the use of modelling and AI techniques to assist the collection, generation and use of evidence as a critical strategic New Approach Method (NAM) methodology supporting SSbD goals. Valuable assistance is obtained from transformative tools that use cutting-edge data and scientific techniques to perform more accurate, ethical, and comprehensive assessments, embedded into SSbD decision making workflows.To achieve this goal, full use of existing data, models and knowledge should be leveraged into providing relevant information supporting assessment and decision goals between alternatives in early stage innovation. Such data needs to be integrated with evidence-weighting and scoring schemes, including uncertainty, to reach initial decisions on alternatives and to plan for subsequent refinement phases. The findings may include recommendations for generating the most meaningful and useful data to address gaps and uncertainty. We will demonstrate the methodology followed and questions arising from current case study work applying SSbD to chemicals, polymers and advanced materials.

Questions to be addressed include:

• How can I substitute a less hazardous ingredient to achieve a required product function?
• What modelling and AI tools and NAMs are available for enabling SSbD?
• How do modelling, AI tools and NAMs assist the generation of knowledge supporting SSbD?
• How do I proceed with my product analysis in initial “data poor” situations?
• What are the environmental impacts of my product design alternatives?
• Can I use life cycle analysis in my early product design and decision making?
• How do I weight information across different impact factors?

Background:

Physical product design is concerned with function, ingredients, formulations, compositions, sourcing, manufacturing and packaging processes, and foreseeing the path of the product along the lifecycle from design through production through consumer use including logistics, use and recycling through to end of life cycle. Data, methods and tools are increasingly becoming available about many of these product design and lifecycle aspects. Exploiting that data for the benefit of reduced health hazards and improved environmental footprints can lead to safer and more sustainable designs of new products, but also greener redesigns of existing products. Knowledge and application of Safe and Sustainable by Design  (SSbD) also helps manufacturers to comply with emerging regulations such as digital product passports, ecodesign and sustainability requireents, and enhancing export opportunities of higher quality products.

The SSbD framework is an innovative strategy that proactively incorporates safety and sustainability into the product development process. By incorporating these concepts from the start of product design, SSbD establishes a new standard for reducing environmental impact and protecting human health, which is critical for innovative and sustainable product design in numerous industries. The increasing growth and variety of businesses that explore alternatives for currently circulating materials that might not reach sustainability standards required by society necessitates an evolution in hazard characterisation techniques as well. Traditional toxicology methodologies frequently rely on substantial animal testing, which poses ethical concerns while also potentially failing to accurately predict real-world human and environmental interactions. 

Reference:

Caldeira, C., Farcal, R., Garmendia Aguirre, I., Mancini, L., Tosches, D., Amelio, A., Rasmussen, K., Rauscher, H., Riego Sintes, J. and Sala, S., Safe and sustainable by design chemicals and materials - Framework for the definition of criteria and evaluation procedure for chemicals and materials, EUR 31100 EN, Publications Office of the European Union, Luxembourg, 2022, ISBN 978-92-76-53280-4, doi:10.2760/404991, JRC128591. https://publications.jrc.ec.europa.eu/repository/handle/JRC128591