The results of combining a numeric extrapolation of data with the methodology of case-based reasoning and expert systems in order to improve the electric load forecasting are presented in this contribution. Registers of power consumption are stored as cases that are retrieved and adapted by an expert system to improve a numeric forecasting given by numeric algorithms. FUTURA software has been developed as a result of this work. It combines the proposed techniques in a modular way while it provides a graphic user interface and access capabilities to existing data bases.