Magnetorheological　(MR)　dampers　are　prominent　semi-active　control　devices　for　vibration　mitigation　of　structures　which　undergo　severe　loads　such　as　earthquakes　or　strong　winds.　However,　because　of　their　inherent　non-linear　characteristics,　development　of　control　strategies　that　can　fully　exploit　the　capabilities　of　MR　dampers　is　an　important　and　challenging　task.　In　this　study,　an　intelligent　control　strategy　integrating　fuzzy　logic　controller　and　Modified　Shuffled　Frog-Leaping　Algorithm　(MSFLA)　is　proposed　to　reduce　seismic　responses　of　structures　installed　with　MR　dampers.　The　MSFLA　is　used　to　adaptively　optimise　the　FLC　parameters　including　Membership　Function　(MF)　parameters,　fuzzy　rules　and　input　scaling　gains.　The　MSFLA-optimised　FLC　is　employed　to　establish　optimal　relationship　between　seismic　responses　and　command　voltages　of　MR　dampers.　Simulation　results　verify　that　the　proposed　controller　outperforms　clipped-optimal　control　and　fuzzy　control　methods.　Also,　the　proposed　controller　can　deal　with　changes　in　structural　parameters　and　earthquake　excitations.　Copyright　©　2016　Inderscience　Enterprises　Ltd.