Planning problems are crucial for providing an efficient use of a flexible manufacturing system (FMS). These include the decisions that have to be taken before the system starts to operate. This study entails joint operation of part selection, operation assignment and tool magazine configuration, which are among the planning problems, and develops a joint mixed integer programming model. Unfortunately, the combinatorial structure of the problem makes the solution difficult and time-consuming for real-world size problems by using the proposed model. Therefore, simulated annealing and tabu search algorithms are developed to solve the problem. The performance of the algorithms is tested on the random-generated problems with different sizes. The tabu search algorithm with long-term memory is found to be the most effective in respect of solution quality.