Evolutionary operators for optimal gate location in liquid composite moulding

A multi-objective evolutionary algorithm is applied to the problem of optimal gate location in liquid composite moulding. The Non-dominated Sorting Genetic Algorithm-II (NSGA-II) is implemented and the efficiency of different variation operators is assessed on a benchmark problem, i.e., a mould of simple geometry. Following this, the algorithm is applied to a more complex mould geometry with the best variation operators. It is shown that the sole use of mutation proved better in the case of a complex part, whereas crossover proved helpful only for a simpler geometry. (C) 2008 Elsevier B. V. All rights reserved.