An approach for developing the optimal operator scheduling solution for a group technology (GT) production problem is studied. A state-transition model is developed to analyse and gain insight into the operator-machine interaction of the problem. Operator cyclic walking patterns are then denned. A Petri net model has succeeded in determining the optimal cyclic walking pattern. The computational efforts needed for the Petri net model are compared with those for an integer programming model. The results show large savings in computational effort by using the Petri net model. In addition, the extendability of the Petri net model for various system aspects is addressed