Due　to　the　rapid　development　of　society　and　economy,　the　large-scale　group　decision　making　(LSGDM)　problems　are　increasingly　common　in　the　real　life.　In　this　paper,　a　dynamic　programming　algorithm-based　picture　fuzzy　clustering　approach　is　proposed　to　solve　the　LSGDM　problems.　First,　aimed　at　the　limitations　of　the　existing　operation　laws　of　picture　fuzzy　sets,　we　define　some　new　picture　fuzzy　operation　laws.　Based　on　these　operation　laws,　a　new　picture　fuzzy　weighted　geometric　operator　is　developed　to　aggregate　the　preference　information　provided　by　the　decision　makers.　Second,　a　new　picture　fuzzy　Dice　similarity　measure　is　proposed　to　detect　the　different　correlations　between　the　decision　makers.　Then,　inspired　by　the　dynamic　programming　algorithm,　a　new　clustering　approach　is　proposed　to　improve　the　efficiency　and　quality　of　the　decision-making.　After　the　clustering　process,　the　picture　fuzzy　score　function　is　employed　to　compare　and　rank　the　alternatives.　Finally,　an　illustrative　example　and　a　comparative　analysis　are　provided　to　demonstrate　the　effectiveness　and　superiority　of　the　proposed　method.　©　2021　Elsevier　Ltd