This　study　innovatively　constructs　an　APF　evaluation　model　integrating　analytic　hierarchy　process　(AHP),　particle　swarm　optimization　(PSO)　and　fuzzy　comprehensive　evaluation　(FCE),　and　integrates　non-dominated　sorting　genetic　algorithm　II　(NSGA-II)　to　construct　a　multi-objective　intelligent　decision-making　framework.　The　index　weight　optimization　is　realized　by　algorithm　coupling,　and　the　evaluation　system　of　tourism　resource　carrying　capacity　including　34　evaluation　dimensions　is　established　by　combining　fuzzy　comprehensive　evaluation.　NSGA-II　is　used　to　solve　multi-objective　collaborative　optimization,　which　breaks　through　the　limitations　of　traditional　single-objective　optimization.　Taking　Hainan　Province　as　a　case　to　verify　the　model,　it　is　found　that　the　overload　degree　of　Sanya　and　Haikou　core　scenic　spots　is　1.04　and　0.98　respectively,　and　it　is　necessary　to　implement　the　tourist　diversion　strategy.　The　surrounding　cities　such　as　Wenchang　(0.82)　and　Wuzhishan　(0.79)　have　the　potential　to　increase　the　carrying　capacity　of　tourists.　Through　the　optimization　model　of　fiscal　expenditure　structure,　under　the　standard　of　140yuan　per　capita　tourism　tax　and　fee,　the　optimal　ratios　of　transportation　infrastructure,　environmental　governance　and　community　construction　are　16　%,　18　%　and　6.8　%　respectively.　The　evaluation-optimization-regulation　full-link　intelligent　decision　support　system　constructed　in　this　study　provides　a　quantifiable　algorithm-driven　solution　for　the　sustainable　development　of　tourism.　©　2025　Copyright　held　by　the　owner/author(s).