As　a　combination　of　statistical　and　mathematical　techniques,　response　surface　models　have　been　recently　found　to　be　capable　of　substituting　FE　models　in　model　updating　iterations　by　using　explicit　mathematical　functions　to　represent　the　relationship　between　the　inputs　and　outputs　of　a　physical　system.　However,　the　literature　related　to　this　topic　is　still　scarce　despite　the　wide　employment　of　the　response　surface　method　in　many　engineering　realms　such　as　chemistry　and　industry.　Due　to　that,　this　paper　attempts　to　propose　a　systematic　damage　assessment　procedure　based　on　the　model　updating　strategy　using　the　response　surface　method.　Instead　of　the　qualitative　evaluation　traditionally　used,　here　the　2(k)　factorial　design　is　employed　to　screen　out　non-significant　updating　parameters　by　quantitative　statistical　analysis,　which　considerably　improves　the　screening　reliability.　Meanwhile,　the　central　composite　design　is　adopted　to　construct　response　surface　models　substituting　original　FE　models　during　updating.　The　proposed　method　is　used　to　detect　the　damage　existing　in　an　experimental　full-scale　bridge.　The　results　demonstrate　the　merits　of　this　method　in　its　easy　implementation　and　high　computation　efficiency,　especially　for　the　bridge　case.