Case　retrieval　is　the　focal　stage　in　a　Case-Based　Reasoning　(CBR)　system.　In　this　paper,　a　new　case　retrieval　method　called　CRGCU　is　proposed　in　a　systematical　way,　which　is　based　on　Genetic　Algorithms　(GAs)　and　Group　Decision-Making　method.　First,　feature　attribute　weights　of　cases　are　optimized　by　using　GAs.　Second,　instead　of　utilizing　only　one　set　of　feature　attribute　weights,　we　calculate　similarity　based　on　multiple　sets　of　optimized　weights.　The　result　of　case　retrieval　can　be　obtained　at　last　by　adopting　Group　Cardinal　Utility　function.　The　experiments　on　both　static　and　dynamic　data　sets　illustrate　that　our　method　is　effective　and　suitable　for　CBR　system.　©2010　IEEE.