Response surface methodology (RSM) is defined as a set of mathematical and statistical techniques that are used to optimize yield of a product. In order to study the effects of urban waste compost and corm rates on flower yield of saffron using RSM, an experiment was conducted at the Research Field of the Ferdowsi University of Mashhad, Iran during the growing season of 2017-2018. The treatments were determined based on low and high levels of urban waste compost (0 and 20 t. ha-1, respectively) and corm weight (5 and 15 t. ha-1, respectively). These were computed by using Minitab V. 17. Dry weight of stigma, dry weight of style, fresh weight of flower, flower number, corm number, dry weight of corm, and number of buds per corm were calculated as dependent variables and changes of these variables were evaluated by using a regression model. Lack-of-fit test was used to evaluate the quality of the fitted model. The adequacy of the model was tested by analysis of variance. In general, the full quadratic polynomial equation was tested to determine significance of the model and the components (such as linear, squared and first-order interaction terms). The quality of the fitted model was judged using the determination coefficient (R2). The results show that the effect of linear component on all traits studied except fresh weight of flower, dry weight of corm 0-4g, and number of buds per corm 0-4 and 8-4 grams were significant. The effect of total grade two had a significant effect on the traits of flower number, fresh weight of flower, dry weight of stigma, dry weight of style, number of daughter corms per 0-4, 4-8 and >8 grams, dry weight of corm per 4-8 and >8 grams and number of buds per >8 grams was significant. Also, the interaction effect of two factors of urban waste compost and weight corm on the number of flowers, fresh weight of flower, dry weight of stigma, dry weight of style, number of daughter corms per 4-8 and >8 grams, dry weight of corm 4-8 and >8 grams, and number of buds per corm >8 grams were significant. Lack of fit test had no significant effect on the studied traits. The full square model for the response variables gave insignificant lack-of-fit indicating that the data of experiment were satisfactorily explained. The highest flower number, fresh weight of flower, stigma dry weight and style dry weight were observed for 5 t. ha-1 corm+ 20 t. ha-1 compost. Optimum of corm weight and urban waste compost and desirability for the traits related to flower and corm with 5 t. ha-1, 20 t. ha-1and d = 0. 92 were urban waste compost with 12 t. ha-1, 20 t. ha-1and d = 0. 95, respectively. Based on the results, municipal waste compost and corm weight had a positive effect on most of the studied traits.