Vector-borne diseases can usually be examined with a vector-host model like the SIRUV model. This, however, depends on parameters that contain detailed information about the mosquito population that we usually do not know. For this reason, in this article, we reduce the SIRUV model to an SIR model with a time-dependent and periodic transmission rate beta(t). Since the living conditions of the mosquitos depend on the local weather conditions, meteorological data sets flow into the model in order to achieve a more realistic behavior. The developed SIR model is adapted to existing data sets of hospitalized dengue cases in Jakarta (Indonesia) and Colombo (Sri Lanka) using numerical optimization based on Pontryagin's maximum principle. A previous data analysis shows that the results of this parameter fit are within a realistic range and thus allow further investigations. Based on this, various simulations are carried out and the prediction quality of the model is examined.