(125b) Using Public Data for Predicting Industrial Chemical Pollution Source Reduction and Prevention Alternatives

Some commercial chemicals have inherent hazardous characteristics that may affect human health
and the environment. Thus, environmental releases and exposure scenarios need proper
identification and assessment. Chemical pollution generation is critical due to the uncertainty of
tracking chemicals across the use and end-of-life supply and management chains. The green
chemistry and engineering principles advise preventing and minimizing pollution generation than
treating and controlling it. Nonetheless, due to several decision-making factors like chemical
properties, environmental regulation, economic behavior, and industrial production size, it could be
difficult to determine a proper source reduction activity/method to prevent hazardous chemical
pollution generation from industrial activities. In addition, the appearance of new chemicals in the
market, new applications for existing chemicals, and data unavailability for decision-making increase
the complexity of selecting adequate strategies for preventing and reducing the impact of chemicals
of concern. This contribution moves forward on creating data engineering for integrating scattered
data with chemical-structured-based machine learning models to predict potential pollution source
reduction activities during the manufacturing of chemicals. This contribution supports environmental
practitioners, stakeholders, and decision-makers in exploring pollution source reduction and
prevention activities and designing strategies for the safer use of chemical substances.