Bridging waste-derived carbon catalyst innovation and machine learning for sustainable biodiesel production: a comprehensive review.

Biodiesel is a promising sustainable alternative to fossil diesel, but its deployment is limited by challenges in catalyst efficiency, stability, reusability, and process optimization. Metal-doped activated carbon (AC) and biochar-based catalysts have attracted considerable attention owing to their high surface area, tunable porosity, and low-cost production from waste biomass. In parallel, machine learning (ML) has emerged as a powerful tool for biodiesel yield prediction, process modeling, and optimization. However, these research areas have largely progressed separately. This review combines scientometric analysis with a systematic review to examine advances in waste-derived AC and biochar-based catalysts and ML applications in biodiesel production. Catalyst synthesis, activation, metal doping, physicochemical properties, catalytic performance, stability, and reusability are critically evaluated alongside ML-based prediction and process optimization approaches. An integrated catalyst-process-performance framework is further discussed to highlight opportunities for incorporating catalyst descriptors into data-driven biodiesel modeling. The review identifies key research trends, knowledge gaps, and future opportunities for advancing sustainable and scalable biodiesel production.