Considering　internal　operation　of　decision-making　unit　(DMU)　while　introducing　technical　heterogeneity　is　the　advantage　of　two-stage　meta-frontier　data　envelopment　analysis　(DEA)　method.　However,　each　DMU　can　obtain　double　shared　ratios　in　different　technical　environments,　and　each　DMU　must　ultimately　determine　one　shared　ratio　under　actual　decision-making,　which　leads　to　a　decision-making　dilemma　of　two-stage　meta-frontier　DEA　in　application.　Therefore,　based　on　traditional　shared　ratio　selection,　some　double　shared　ratio　selection　methods　of　dominant,　autonomous　and　coordinated　are　proposed　for　different　decision-making　needs,　which　promote　deep　integration　of　meta-frontier　and　two-stage　DEA,　and　thus　provide　clear　and　specific　shared　ratio　selections　for　decision　makers.　In　addition,　to　overcome　the　limitations　of　double　shared　ratio　selection,　this　paper　takes　double　shared　ratios　as　a　selection　interval　and　further　extends　double　shared　ratio　selection　methods,　enhancing　the　flexibility　of　shared　ratio　selection.　Furthermore,　different　double　shared　ratio　selections　are　divided　into　uniform　and　non-uniform　for　comparative　analysis.　This　paper　indicates　that:　(1)　The　degree　of　concentration　in　DMU　shared　ratio　selection　is　positively　correlated　with　efficiency.　(2)　Unifying　the　shared　ratios　for　meta-frontier　and　group　frontier　is　beneficial　for　improving　overall　efficiency　level.　©　The　Author(s)　under　exclusive　licence　to　Sociedade　Brasileira　de　Matemática　Aplicada　e　Computacional　2025.