This paper presents the use of particle swarm optimization to maximize the clean coal yield of a coal preparation plant that typically has multiple cleaning circuits that produce the same product quality so that the blend of clean coal meets the targeted product quality contraints. Particle swarm is used for the yield optimization while satisfying multiple product quality restraints. The results show a 2.73% increase in the yield can be achieved leading to additional revenue of $5,460,000 per annum for a plant producing 10 million tons of clean coal per year without significantly adding to the implementation/operation cost.