Abstract: From the perspective of synergetic of system management, the structure and objectives of ordering system are analyzed. Then, a flexible constrained order allocation model is established to improve the overall performance of this system. In this model, an enterprise's elasticity of delivery time, product quality and other aspects about the ordering system are added, which extends the "hard" constraints in traditional model to fuzzy "soft" constraints. The model uses fuzzy membership functions to reflect the degrees of an enterprise's satisfaction toward various system factors. Examples show that the proposed model not only maintains the advantages of traditional programming models, but more importantly, it could optimize the main goal of the system and the secondary goals collaboratively. Meanwhile, the outputs of the model are robust to change of parameters, which is a good reflection of the advantage generated by the overall optimization in the model.

Key words: fuzzy system optimization, fuzzy programming, order allocation, triangular fuzzy numbers