An　innovative　methodology　for　the　probabilistic　estimation　of　ground　condition　and　construction　cost　is　proposed　in　this　article,　which　is　an　integration　of　the　geology　prediction　approach　based　on　Markov　process　and　the　cost　variance　analysis　based　on　work　breakdown　structure.　The　former　one　provides　the　probabilistic　description　of　ground　classification　along　tunnel　alignment　according　to　the　geology　information　revealed　from　boring　holes.　The　latter　one　provides　the　probabilistic　description　of　expected　cost　for　each　interested　operation　according　to　the　survey　feedbacks　among　practitioners.　Then　an　engineering　application　to　Chuanshi　Tunnel　is　well　presented　to　demonstrate　how　the　ground　condition　and　the　construction　cost　are　estimated　in　a　quantitative　and　probabilistic　way.　It　facilitates　both　the　owners　and　the　contractors　to　be　aware　of　the　risk　they　should　carry　before　construction,　and　it　is　meaningful　for　both　tendering　and　bidding.　©　2014　Elsevier　Ltd.