In　project　management,　it　is　an　important　approach　that　the　deep　excavation　safety　risk　(DESR)　is　synthetically　assessed　based　on　experiential　information　provided　by　experts.　This　information　is　usually　obtained　from　multiple　sources　and　is　incomplete.　Therefore,　the　problem　of　synthesizing　multiple　sources　and　uncertainty　information　is　of　great　significance　for　the　assessment　and　control　of　DESR.　This　paper　presents　a　risk　assessment　and　decision　approach　for　deep　excavation　construction　based　on　Bayesian　analysis.　First,　the　expressions　of　a　likelihood　function　used　to　describe　and　synthesize　uncertain　information　obtained　from　expert　panel　is　constructed,　and　a　discretization　interval　division　method　of　occurrence　probability　and　consequences　of　an　unwanted　event　is　proposed　based　on　the　current　risk　code　in　China.　Then,　the　discrete　model　of　synthesizing　information　of　expert　estimates　is　constructed,　and　a　comparative　analysis　of　expert-elicited　results　and　formula　deduction　are　utilized　to　evaluate　the　advantages　of　this　model　under　some　hypothetical　scenarios.　On　that　basis,　the　index　system　of　DESR　is　established　and　the　weight　of　each　index　is　calculated　by　the　method　of　analytic　hierarchy　process.　The　risk　matrix,　deviation　degree　and　risk　acceptance　criteria　are　developed　to　determine　the　whole　risk　level.　Finally,　the　proposed　method　is　verified　by　analyzing　a　typical　deep　excavation　of　Beijing　Metro.　The　results　demonstrate　the　feasibility　of　this　method　and　its　application　potential.　©　2017,　The　Society　for　Reliability　Engineering,　Quality　and　Operations　Management　(SREQOM),　India　and　The　Division　of　Operation　and　Maintenance,　Lulea　University　of　Technology,　Sweden.