论文标题
使用交互式进化算法的多物理组合优化：设施位置问题的情况
Multiobjective Combinatorial Optimization with Interactive Evolutionary Algorithms: the case of facility location problems
论文作者
论文摘要
我们考虑通过偏好驱动有效的启发式方法处理的多主体组合优化问题。他们根据决策者在此过程中表达的一些偏好来寻找帕累托阵线中最喜欢的部分。通常，在这种情况下搜索的是帕累托的有效解决方案集。这比优化单个目标函数要困难得多。此外，获得帕累托集并不意味着解决决策问题，因为必须选择一种或某些解决方案。确实，要做出决定，有必要确定帕累托集中最喜欢的解决方案，以便也有必要引起用户的偏好。从这个角度来看，我们提出的内容可以看作是设施位置问题中的第一个结构化方法，以考虑到用户的偏好，以搜索最佳解决方案。以此目的，我们使用最近提出的称为Nemo-II-CH的互动进化的多目标优化程序来解决设施的位置问题。 Nemo-II-CH适用于许多用户和许多设施的现实世界多目标位置问题。已经执行了一些虚拟用户的几个模拟。将NEMO-II-CH获得的结果与通过三种算法获得的结果进行了比较，这些算法知道用户的真实价值函数，而Nemo-II-CH不知道。他们表明，在许多情况下，NEMO-II-CH比确切知道整个用户的真实偏好的方法更快地找到了位置的最佳子集。
We consider multiobjective combinatorial optimization problems handled by means of preference driven efficient heuristics. They look for the most preferred part of the Pareto front on the basis of some preferences expressed by the Decision Maker during the process. In general, what is searched for in this case is the Pareto set of efficient solutions. This is a problem much more difficult than optimizing a single objective function. Moreover, obtaining the Pareto set does not mean that the decision problem is solved since one or some of the solutions have to be chosen. Indeed, to make a decision, it is necessary to determine the most preferred solution in the Pareto set, so that it is also necessary to elicit the preferences of the user. In this perspective, what we are proposing can be seen as the first structured methodology in facility location problems to search optimal solutions taking into account preferences of the user. With this aim, we approach facility location problems using a recently proposed interactive evolutionary multiobjective optimization procedure called NEMO-II-Ch. NEMO-II-Ch is applied to a real world multiobjective location problem with many users and many facilities to be located. Several simulations considering different fictitious users have been performed. The results obtained by NEMO-II-Ch are compared with those got by three algorithms which know the user's true value function that is, instead, unknown to NEMO-II-Ch. They show that in many cases NEMO-II-Ch finds the best subset of locations more quickly than the methods knowing, exactly, the whole user's true preferences.
