Material recycling contributes not only material circulation but also reduction of greenhouse gas (GHG) emissions since GHG emissions for a virgin material production can be saved by using recycled materials. Partial disassembly is required to maximize GHG saving rate defined as the saved volumes of GHG emissions by recycling, and recycling rate, and to minimize cost. This study proposes a multi-objective disassembly parts selection for cost, recycling and GHG saving rates using Linear Physical Programming (LPP). The LPP is one of the effective methods for solving multi-objective problems, and can obtain one satisfactory solution by expressing preferences of the decision maker (DM) for each objective. The proposed model was validated using a vacuum cleaner consisted of 23 parts, and found one satisfactory solution, which had over 70.00[%] and 80.36[%] of recycling and GHG saving rates and achieved over 50% reduction of total cost compared to complete disassembly.