Current　life-cycle　sustainability　assessment　(LCSA)　methodology　for　pavement　alternatives　is　deterministic,　which　generally　reduces　the　reliability　of　the　results.　This　study　proposed　an　uncertainty-based　LCSA　framework　for　pavement　alternatives　to　capture　the　uncertainties　associated　with　economic,　environmental,　and　social　pillars　during　the　life　cycle.　After　identifying　and　characterizing　the　uncertainties,　the　probabilistic　LCSA　results　associated　with　eight　impact　categories　were　obtained　using　Monte　Carlo　simulation.　Subsequently,　an　improved　multi-criteria　decision-making　approach　was　presented　to　unify　the　random　outcomes　of　the　eight　impact　categories,　where　the　probabilistic　integrated　sustainability　degrees　(ISDs)　were　obtained　to　represent　the　sustainability　level　for　the　alternatives.　Meanwhile,　the　probability　of　being　best　option　for　each　alternative　was　obtained.　A　case　study　was　applied　to　perform　the　uncertainty　analysis　of　LCSA　including　hot　mix　asphalt　(HMA)　pavement,　warm　mix　asphalt　(WMA)　pavement,　and　reclaimed　asphalt　pavement　(RAP).　The　results　showed　the　economic　performance　of　RAP　alternative　was　significantly　less　than　that　of　HMA　and　WMA　alternatives　at　confidence　levels　of　97%　and　99%,　respectively.　Moreover,　there　were　no　statistically　significant　differences　among　three　alternatives　with　respect　to　relevant　environmental　impact　categories.　Three　pavement　alternatives　denoted　positive　social　impacts　with　95%　confidence　interval.　In　addition,　the　satisfactory　ISD　was　obtained　for　RAP　alternative　with　95%　confidence　interval,　while　RAP　alternative　was　selected　as　the　best　option　with　a　high　probability　of　91.71%.　In　addition,　it　was　found　that　the　ISDs　were　sensitive　to　the　economic　and　social　weights　in　the　case　study.　This　study　extends　the　existing　pavement　LCSA　framework　by　capturing　the　multivariate　potential　uncertainties,　and　thus　represents　a　valuable,　reliable　tool　for　decision　makers.　©　2020　Elsevier　Ltd