In　this　paper,　a　damage　localization　method　based　on　decision　tree　is　proposed,　which　is　in　particular　used　in　localization　of　offshore　jacket　platforms　to　improve　its　identification　accuracy　and　robustness.　Modal　shape　change,　modal　slope　change　and　modal　curvature　change,　as　well　as　decision　tree　attributes,　are　calculated　based　on　the　first　modal　shape　of　the　structure.　Several　damage　cases　from　an　offshore　jacket　platform　model　are　localized　to　illustrate　the　proposed　method　finally.　The　numerical　example　shows　that　the　IA　of　damage　localization　decreases　with　the　lower　extent　damage　and　higher　noise　level.　Experimental　results　show　that　the　identification　accuracy　is　over　90%　for　the　jacket　platform.　This　indicates　that　the　proposed　method　is　efficient　for　structural　damage　detection;　furthermore,　it　also　has　good　applicability　and　practicality　to　some　extent　in　practical　engineering.