While people need to know tomorrow’s weather to decide suitable activities and precautions, so do the “intelligent” building management systems. The accuracy of the short-term prediction of the ambient conditions is particularly import for the development of predictive control strategies. Although the shortterm prediction methods for outside air temperature have been extensively studied, reliable prediction methods for solar radiation are yet to be established. This paper discusses the possibility of using meteorological forecast data with local observation data for short-term prediction of the ambient conditions. Based on the analysis of the collected data, a new method has been developed for temperature and solar radiation prediction. It is found that a linear combination of the observation and the forecast data can effectively improve the accuracy of short-term (24 hours) prediction. The method is also suitable for the prediction of humidity and wind speed/direction.