A perspective in data and software systems to streamline the transition towards a safer chemical end-of-life management

Chemicals serve a pivotal role in many products. To manage their environmental impact, chemical risk assessment (CRA) and material flow analysis (MFA) are used. However, data gaps, particularly at the end-of-life (EoL) stage, hinder these efforts. This paper explores how software and data systems can support CRA and MFA by integrating regulatory databases, extracting information from academic sources via natural language processing, and incorporating real-time data. These advances improve understanding of the EoL supply and management chain via data integration and automated tracking. Graph neural networks and transfer learning are applied to enhance model representation and predictive performance.