Multi-Objective Reaction Optimization: Yield vs. Waste Pareto Front
Watch a multi-objective genetic algorithm evolve a population of temperature/catalyst reaction conditions across generations, splitting them into non-dominated Pareto fronts that trade reaction yield against process waste (E-factor).
A population of candidate reaction conditions — temperature and catalyst loading pairs — sits on a 3D response-surface landscape shaped like real reaction yield, and evolves generation by generation under a multi-objective genetic algorithm. Every point is scored on two objectives at once: how much product it yields, and how much waste (E-factor) it generates getting there. Non-dominated sorting classifies the population into fronts each generation — the gold points are the Pareto front, conditions where you cannot improve yield without generating more waste, or cut waste without losing yield. Step through generations manually or auto-run the search, watch the front tighten and migrate across the landscape, and read the live Pareto-front size and its best yield / lowest-waste extremes.
Control temperature and catalyst levels to optimize yield while minimizing waste in chemical reactions.
3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install