Abstract The new approaches to the modelling of the Sun-solar wind-magnetosphere-radiation belts chain and their implementation at MSU's Space weather center (Space Monitoring Data Center, SMDC) are presented. The main idea is to use data from medium-term (3–5 days) forecasts of the speed of quasi-stationary solar wind streams based on data obtained from solar images in the UV wavelength range to improve the quality of forecasts of the magnetospheric factors of the space environment. An operational model for predicting the daily fluence of relativistic electrons (>2 MeV) in the Earth's outer radiation belt has been developed, with the extended set of input parameters that include the forecast of the solar wind velocity. To validate the model, the period from October 11, 2016 to February 18, 2017 was considered, during which the Sun made several revolutions and it was possible to observe the evolution of coronal holes and the development of solar sporadic phenomena. The arrival of high-speed solar wind streams from coronal holes to the Earth's magnetosphere, as well as, as impact of coronal mass ejections, provoked geomagnetic disturbances of various intensity, and caused significant variations of the electron fluxes of the Earth's outer radiation belt. SMDC website implements solar wind model based on the data of solar observations by the AIA instrument of the SDO space observatory at wavelengths of 19.3 and 21.1 nm that forecasts the daily fluence of relativistic electrons (E > 2 MeV) in geostationary orbit. It is shown that the use of the results of medium-term forecasting of the solar wind speed as an input parameter in the daily electron fluence forecast model can significantly improve its quality with a forecast horizon of up to 4 days.