The constant increase in the requirement for electricity, the limited availability of resources and the lack of storage of electricity force the sector participants to prepare various action plans and safety measures. The necessity of consuming electricity at the moment it is produced imposes the provision of supply-demand balance to market actors.

The aim of this study is to enhance a time series model that estimates energy demand. It is possible to obtain energy demand forecasts in the short, medium and long term using statistical methods and artificial intelligence algorithms.

In order to forecast energy demand , weather, seasonality, GDP, population, import, export, area of the building and number of vehicles data are used as input of the model.

The fact that electricity cannot be stored is increasing the importance of planning the supply-demand balance and estimation of the power generation with high accuracy. The accuracy of the estimations will also increase the validity of the planning.

Forecasting errors lead to unbalanced supply-demand, which adversely affects the operational cost, reliability and efficiency. For this reason, prediction accuracy plays an important role.

Strong inferences can be obtained by minimizing the error with statistical methods and artificial intelligence algorithms. Forecasting energy production is usually based upon meteorological data like solar irradiation, temperature.

With IBM's leading rules engine Decision Manager we address complex contract pricing demands. Weather it's a retail client or an industrial enterprise you can count on pricing decisions that's governed by your business analyst.

Dynamic pricing, segment based pricing, hour based pricing, consumption segmentation is easily regulated and integrated with you SAP platform with JPricing for Utilities solution.