To　reasonably　and　effectively　evaluate　the　stability　of　jointed　rock　slopes,　this　study　proposes　a　reliability　analysis　method　that　considers　the　uncertainty　of　the　joint　spatial　distribution.　First,　the　discrete　fracture　network　(DFN)　model　is　used　to　characterize　the　variability　of　joint　geometric　parameters,　and　a　random　variable　of　joint　location　is　assumed　to　be　a　random　integer　that　follows　a　uniform　distribution　in　the　random　number　library.　Then,　Latin　hypercube　sampling　is　used　to　sample　uncertain　random　variables　to　generate　random　numbers,　and　construct　the　corresponding　DFN　model.　Finally,　the　safety　factor　(Fs)　of　each　numerical　model　under　sampling　is　calculated　by　strength　reduction　method,　and　the　failure　probability　of　jointed　rock　slope　is　obtained　by　statistical　analysis　of　Fs.　The　comparative　analysis　results　demonstrate　that　considering　the　uncertainty　of　joint　spatial　distribution　is　crucial　for　accurately　analyzing　the　reliability　of　rock　slopes.　This　method　comprehensively　considers　the　main　internal　factors　that　affect　the　reliability　of　jointed　rock　slopes,　including　the　uncertainty　of　rock　strength　parameters　and　variability　of　geometrical　parameter　and　the　location　randomness　of　joints,　making　the　results　of　the　evaluation　of　the　reliability　of　rock　slopes　more　precise　and　reasonable.　©　2023