You could be looking at maximizing the distances between the different clusters (high inter-cluster distances) while having each cluster as compact as possible (low intra-cluster distances).
Check the wikipedia page about cluster analysis: https://en.wikipedia.org/wiki/Cluster_analysis
Mr. Amir Ehsan Ranginkaman, The role of clustering is the task of grouping population, in such a way the members of each group are similar finally. what are fitness functions in multi objective algorithms when they combine with clustering method? Are they related to objectives of optimization problem or they are related to minimizing the distance between clusters members and center of clusters?!
As a follow up to your comment, it could be related to just clustering as in Brain Storm Optimization (BSO) where similar individuals (representing ideas) are grouped in clusters in every iteration.
Mr. Amir Ehsan Ranginkaman I just want to use clustering method to form several sub swarms with the MOPSO algorithm in order to thoroughly sample the search space. You are right. I agree with you and think fitness functions in MOPSO algorithm depend on optimization problem.