Carbon dioxide is an essential component of the Earth’s atmosphere. Due to high energy consumption, residential and office buildings are responsible for one-third of the total CO2 emissions in the European Union [EU19]. The energy industry has the highest contribution of the Greenhouse Gases (GHGs) emissions in Germany [uba19]. Carbon dioxide is considered the most harmful greenhouse gas in the atmosphere regarding its effect on global warming and climate change issues [NASA19]. The amount of CO2 emitted to produce electricity (i.e., CO2-equivalent intensity factor) can greatly vary in time, depending on the sources used to generate it. CO2-efficient rescheduling of electricity consumption and integration of energy carriers may enable up to 40% emission reductions [FA18]. The thesis proposes an implementation of a service-oriented architecture that provides one-day ahead forecasts for the CO2-equivalent intensity factor in Germany using electricity generation forecasting data from ENTSO-E API [ENTSO19] and Weather forecasting data from Darksky API [darksky19]. This work proposes an alternative approach to choose and engineer the required independent variables for the forecasting process. Ordinary Least Squares, Ridge Regression, Polynomial Regression, Decision Trees, Random Forest, and KNNs are applied and evaluated to model the forecasting problem. The results show that the Random Forest model has better evaluation performance among the six models with a R2 value of 0.94 on a separated test dataset. Therefore, the implemented service-oriented architecture uses the Random Forest model to forecast the CO2-equivalent intensity factor in Germany. The implemented architecture provides the forecasts using a REST API based architecture through an HTTP GET request with a JSON response.