Establishing　a　suitable　TBM　performance　prediction　model　is　benefit　to　the　machine　type　selection,　project　scheduling　and　budgeting.　Some　regression　equations　using　rock　mass　classification　parameters　had　already　been　set　up　based　on　TBM　tunneling　database,　some　of　which　were　small.　Besides,　most　of　existing　equations　involved　field　penetration　index　(FPI)　or　the　cutter　force.　Once　apply　these　equations,　how　to　determine　the　cutter　load　is　a　problem.　The　cutter　thrust　force　is　affected　by　various　factors,　the　most　important　of　which　is　the　ground　conditions.　To　make　a　straightforward　estimation　of　penetration　rate　at　planning　phase,　some　simple　regression　equations　merely　using　rock　mass　classification　parameters　were　set　up　based　on　115　tunnel　sections　sourced　from　three　TBM　job　sites.　TBM　penetration　rate　(mm/rev)　was　employed　as　the　prediction　result.　It　is　a　comprehensive　value　including　the　rock　mass　condition　and　the　experiences　of　TBM　operators　in　the　projects　used　for　this　study,　since　TBM　operators　would　adjust　the　thrust　force　with　the　changing　rock　masses　to　get　an　acceptable　penetration　rate　on　site.　The　datasets　in　the　database　indicated　the　intrinsic　relationship　among　the　rock　mass　condition,　the　needed　thrust　force　and　the　obtained　penetration　rate.　Five　rock　mass　classification　systems　including　RMR,　Q,　GSI,　HC　and　BQ　were　covered,　the　latter　two　are　widely　used　in　China.　The　input　rock　mass　classification　parameters　were　categorized　into　three　types,　which　respectively　are　final　ratings,　estimated　rock　mass　properties,　and　partial　ratings　or　corresponding　original　values.　Generally,　the　multivariate　equations　using　partial　ratings　have　the　best　correlation　with　TBM　penetration　rate.　The　strength　of　intact　rock,　integrity　of　rock　mass　and　joint　condition　are　proved　to　be　the　most　critical　factors　in　the　estimation　of　TBM　penetration　rate.　The　established　equations　reflect　the　statistical　relationship　between　the　rock　mass　and　penetration　rate　with　the　requisite　of　the　normal　TBM　tunneling　data　in　various　rock　masses,　and　can　be　easily　applied　in　the　planning　phase　of　a　TBM　tunneling　project.