Considering　the　influence　of　pipe　aging　and　environmental　factors,　the　failure　probability　assessment　of　in-service　water　supply　pipes　can　guide　the　renovation　and　maintenance　of　water　supply　pipes.　Taking　pipe　diameter,　pipe　age,　pipe　length,　ground　type,　buried　depth　and　inner　pressure　as　input　variables,　this　paper　uses　Bayesian　network　(BN)　to　establish　a　water　supply　pipe　failure　event　classification　model　to　predict　the　failure　probability　of　water　supply　pipes.　The　proposed　methodology　is　illustrated　with　a　real-world　case　of　water　supply　system　in　China.　The　application　results　shows　that　the　BN　model　has　a　good　prediction　efficiency,　which　can　be　used　to　estimate　the　health　status　of　pipes　and　identify　the　high-risk　pipes　in　the　pipe　network　based　on　the　prediction　results,　so　as　to　provide　decision　support　for　the　renewal　and　renovation　of　water　supply　pipes.　©　Conference　Proceedings　of　the　10th　International　Symposium　on　Project　Management,　China,　ISPM　2022.