Heavy rainfall events during the Indian summer monsoon (ISM) have increased dramatically in recent years. These events cause landslides and flash floods that result in a significant loss of life and property every year during the Indian monsoon. An accurate prediction of the high-intensity precipitation events during the ISM is critical. The accuracy of a numerical weather prediction (NWP) model forecast depends on the accuracy with which the model represents the physics of the atmosphere and the exactness of the initial conditions provided to the model. Due to the limitation in the amount of computational power, there is a limit on the resolution at which the NWP models make a prediction. The processes that occur at a sub-grid scale are parameterized. These parameterization schemes contain tens of parameters, whose values are assigned either theoretically or by experimental investigations from the scheme developers. The values of these parameters significantly influence the accuracy of the prediction. Therefore, it is critical to calibrate the values of these parameters to obtain an accurate prediction. A multiobjective adaptive surrogate model-based optimization (MO-ASMO) method is used to calibrate nine sensitive parameters that greatly influence the model prediction. Twelve high-intensity four-day precipitation events of ISM during the years 2015-2017 over the monsoon core region in India are considered to calibrate the model parameters. Normalized root-mean-square error (NRMSE) values corresponding to four meteorological variables precipitation, surface air temperature, surface air pressure, and wind speed are minimized by calibrating the sensitive parameters. The exactness of the initial conditions provided to an NWP model also impacts the accuracy of the forecast. The impact of conventional data and satellite radiance data assimilation on the short-range prediction of the ISM is inspected. Three-dimensional ensemble variational (3DEnVar) hybrid method is used for assimilating the observations. The impact of assimilation on the NWP model with the default and calibrated parameters is examined.