Abstract: The purpose of this research was to apply the optimization method by the specific technique of Grey Wolf Optimizer (GWO) cooperating with the reservoir simulation model in order to improve the rule curves of reservoirs. The minimal average water shortage, the minimal average excess water, the minimal frequency of water shortages, the minimal frequency of excess water were used as the objective functions for searching procedure. This study considered the monthly data of reservoir rule curves from the Ubolratana reservoir, Khon Kaen Province, the Lampao reservoir, Kalasin Province and the Nam Oon reservoir, Sakon Nakhon Province as the case studies. Moreover, this study synthesized 1,000 samples of inflow data in order to evaluate the efficiency of new rule curves from the proposed model. The results were displayed the situations of water shortage and excess water in term of frequency, duration time, average water quantity and the highest water quantity. The results shown that four objective functions above can produce new rule curves which their physicals were similar to the current ones for all reservoirs due to the influences of seasonal inflow and the same operation conditions. The results after testing the new rule curves compared with the current ones show that the new rule curves improved by the objective function of minimal average water shortage can mitigate excess water and water shortage situations better than other rule curves produced by other objective functions and also better than current rule curves for all reservoirs. The results can conclude that GWO with the objective function of minimal average water shortage connecting with reservoir simulation model has more efficiency to search optimal rule curves.