This　paper　proposes　a　stiffness　identification　method　of　timber　joints　via　generalized　likelihood　ratio　and　quality　control　chart.　Firstly,　the　root-mean-square　(RMS)　of　generalized　likelihood　ratio　is　analyzed　for　the　collected　structural　responses　of　timber　joints,　simultaneously　the　maximum　and　minimum　of　the　first　principal　component　are　extracted　and　the　quality　control　chart　is　drawn.　Secondly,　the　extents　of　damage　are　determined,　which　are　based　on　the　degree　of　the　maximum　and　minimum　values　of　the　first　principal　component　scores　over　control　boundaries.　Finally,　the　quasi-static　test　of　a　timber　joint　is　completed　to　verify　the　proposed　approach.　The　results　show　that　an　exponential　relation　is　existed　between　the　stiffness　degradation　ratio　of　timber　joints　and　the　extent　of　control　indicators　over　the　upper　and　lower　boundaries,　and　a　stiffness　predicted　formula　is　predicted　that　can　be　able　to　predict　the　stiffness　degradation.　©　2015,　International　Society　for　Structural　Health　Monitoring　of　Intelligent　Infrastructure,　ISHMII.　All　rights　reserved.