Sequence Sort: A new non-dominated sortingalgorithm for evolutionary multi-objective optimization

Abstract Non-dominated sorting is a crucial component of multi-objective evolutionary algorithms. However, efficient and straightforward Pareto-based non-dominated sorting algorithms remain scarce. This paper presents a new non-dominated sorting strategy, called Sequence Sort, which achieves a best-case computational complexity of O(MN√N), where N is the population size and M is the number of objectives. Sequence Sort adopts strategies combining presorting and solution marking. Comparative experiments with four mainstream non-dominated sorting algorithms demonstrate that Sequence Sort offers a clear advantage in computational efficiency, providing a more effective and reliable approach for solving multi-objective optimization problems.