A novel approach in selective assembly with an arbitrary distribution to minimize clearance variation using evolutionary algorithms: a comparative study

The minimization of surplus components with normal dimensional distributions while making selective assemblies was the only objective considered in the previous research works carried out by various researchers in different periods. Seldom works have been found on selective assembly by considering all dimensional distributions. In this proposed work, a novel method is developed for making assemblies with zero surplus components and minimum clearance variation by considering arbitrary distribution, to demonstrate the greater improvement in the results than the past literature. Krill Herd algorithm has been implemented for identifying the best combination of groups. Computational results showed that the proposed krill herd algorithm outperformed as compared with existing literature and as well as the results by gaining-sharing knowledge-based algorithm, differential evolution algorithm, and particle swarm optimization algorithm.