Modeling and optimization trends in agricultural waste pyrolysis for heavy metals removal from wastewater

The need for cost-effective and sustainable treatment methods has increased due to the growing build-up of heavy metals in wastewater. Agricultural wastes that have been pyrolyzed have become important biochar precursors, offering inexpensive adsorbents with large surface areas, porous structures, and lots of functional groups for the removal of heavy metals. The modeling and optimization techniques used to remove heavy metals from wastewater and leachate using pyrolyzed agricultural wastes are the main topic of this comprehensive review of the literature. After searching Scopus and Web of Science using the PRISMA framework, 47 pertinent studies were found out of 626 records (2016–2026). The ideal pyrolysis temperatures are usually between 400 and 800 degrees Celsius. Common feedstocks include rice husk, wheat straw, coffee grounds, corn cobs, and banana pseudostem. For process optimization, traditional statistical techniques are still the most popular; nonetheless, this study highlights the potential uses of machine learning models. Additionally, the paper proposes the potential application of deep learning architectures for precise adsorption efficiency prediction and mechanistic understanding. Hence, future research should prioritize interpretable, ensemble, deep learning, small language models, and multi-agents to guide biochar design and operational optimization for heavy metal removal from wastewater.