In this paper, we introduce a new technology permanently installed on the well completion and addressed to a real time reservoir fluid mapping through time-lapse electric/electromagnetic tomography while producing and/or injecting. Our technology consists of electrodes and coils installed on the casing/liner in the borehole/reservoir section of the well. We measure the variations of the electromagnetic fields caused by changes of the fluid distribution in a wide range of distances from the well, from few meters up to hundreds meters. The data acquired by our technology are processed and interpreted through an integrated software platform that combines 3D and 4D geophysical data inversion with a Machine Learning platform equipped with a complete suite of classification/prediction algorithms. Every time new data are acquired, they are fully integrated with the previous database, and used for decreasing the level of uncertainty about the dynamic model of the reservoir. In order to clarify the potential impact of such system on reservoir management, we apply this methodology to a synthetic data set. We discuss a simulation of a scenario where the waterfront approaches the wells during oil production. The goal of our test is to show how to combine our technology with Machine Learning to make robust predictions about the water table variations around the production wells.