A flexible manufacturing system (FMS) is an integrated system of computer numerically controlled machine tools connected with automated material handling. A set of production planning problems for FMSs has been defined (Stecke [Stecke, Kathryn E. 1983. Formulation and solution of nonlinear integer production planning problems for flexible manufacturing systems. Management Sci. 29 (3, March) 273–288.]), and this paper considers one called the loading problem. This problem involves assigning to the machine tools, operations and associated cutting tools required for part types that have been selected to be produced simultaneously. The part types will be machined during the upcoming production period (say, of one to three weeks duration on average) and according to a prespecified part mix. This assignment is constrained by the capacity of each machine's tool magazine as well as by the production capacities of both the system, and each machine type. There are several loading objectives that are applicable in a flexible manufacturing situation. This paper considers the most commonly applied one, that of balancing the workload on all machines. This paper first discusses a nonlinear integer mathematical programming formulation of the loading problem. The problem is formulated in all detail. Then an efficient solution procedure is proposed and illustrated with an example. Computational results are provided to demonstrate the efficiency of the suggested special-purpose procedures.