Review on Predictive Models and Integration Strategies for Holistic Impact Assessment of Chemicals and Materials

Abstract

Rapid innovation in chemicals and materials calls for innovative integrated approaches that can assess their impacts across different areas. The Safe and Sustainable-by-Design (SSbD) framework, developed by the European Commission's Joint Research Centre (JRC), offers a comprehensive approach with which to evaluate the safety and sustainability of chemicals and materials across their lifecycle. While SSbD uses various modeling approaches to assess impacts on human health, the environment, and socioeconomic factors, these are often applied independently, hindering a holistic understanding of the complex interactions between these factors and thus the simultaneous optimization of function, cost, safety and sustainability. This review describes existing predictive models and available strategies for their integration to facilitate more comprehensive and holistic chemical and material impact assessments. Specifically, we examine three model integration strategies: consensus integration that combines model predictions for the same impact categories, weighted aggregation that combines different scores in a unified one, and pipeline integration that links models sequentially to create a more unified assessment. Furthermore, we address key concepts related to the uncertainty of model predictions and the applicability domain of models, highlighting how these evolve in integrated frameworks. Insights into the applications of these integration strategies and challenges will allow a more accurate, coherent, and sustainable approach to chemical and material safety and sustainability assessments.