This paper presents an integer programming model for existing FMS users to select the most cost effective set of parts to run simultaneously on an FMS during a specified production horizon. Two heuristic solution algorithms are developed by dividing the part-selection procedure in two stages. One stage deals strictly with the limitations on machining time, storage capacity, and AGV time. For the other stage, three different stages are used to choose a set of pan solely based on tooling and fixture constraints. The best algorithm-strategy combination has been identified and proved to be very efficient in finding optimal or near optimal solutions.