Application of a hybrid multi-objective evolutionary algorithm to the uncapacitated exam proximity problem
Application of a hybrid multi-objective evolutionary algorithm to the uncapacitated exam proximity problem
复制标题
DOI：
--
复制
发表时间：
2004
期刊：
影响因子：
--
通讯作者：
P. Côté;Tony Wong;R. Sabourin
中科院分区：
文献类型：
--
作者：
P. Côté;Tony Wong;R. Sabourin
混合多目标进化算法用于解决无能力考试邻近问题。在这种杂交中，使用局部搜索算子代替传统的遗传重组算子。搜索算子之一旨在修复由初始化过程和变异算子产生的不可行的时间表。另一个搜索算子实现了简化的可变邻域下降元启发式，其作用是提高邻近成本。将所得的非支配时间表与使用 15 个公共领域数据集的其他优化方法产生的时间表进行比较。无需特殊微调，混合算法就能够生成在 15 个数据集中的 9 个数据集中排名第一和第二的时间表。
A hybrid Multi-Objective Evolutionary Algorithm is used to tackle the uncapacitated exam proximity problem. In this hybridization, local search operators are used instead of the traditional genetic recombination operators. One of the search operators is designed to repair unfeasible timetables produced by the initialization procedure and the mutation operator. The other search operator implements a simplified Variable Neighborhood Descent meta-heuristic and its role is to improve the proximity cost. The resulting non dominated timetables are compared with thouse produced by other optimization methods using 15 public domain datasets. Without special fine-tuning, the hybrid algorithm was able to produce timetables ranking first and second in 9 of the 15 datasets.