During　the　partner　selection　process　in　a　virtual　enterprise,　the　information　about　the　candidates　and　their　performances　are　incomplete　and　uncertain.　In　this　paper,　vague　sets　theory　is　introduced　to　give　comprehensive　consideration　to　truth　membership,　false　membership　and　uncertainty　membership　degree　of　the　candidates　to　satisfy　the　decision-maker.　On　the　basis　of　the　agreement　index　of　satisfaction　degree,　the　formulated　partner　selection　problems　are　interpreted　so　as　to　maximize　the　average　agreement　index.　To　solve　the　problem,　genetic　algorithm　is　proposed.　Finally,　the　simulation　of　a　case　demonstrates　that　the　method　is　effective.　©　2010　IEEE.