The relatively low average conversion efficiency of air-conditioning systems and the recently imposed upper bounds to the final energy use in the heating and cooling of residential buildings suggest to consider new approaches to design less energy intensive systems. An integrated, exergy-based approach for the optimal matching of internal and external heating plants in building conditioning systems has been proposed — and its theoretical basis discussed — in a previous paper. The procedure allows the designer to obtain a pseudo-optimal integration of the building and its heating plant (heating element + primary energy supply system) and to identify, among a set of alternative solutions for the building under examination, the thermodynamically most efficient plant. The objective of this paper is to validate the method on a real building in order to demonstrate its practical applicability. The large “Chiostro Hall” (220 square meters, 1245 cubic meters) of the Engineering School of the University “Sapienza” of Roma has been employed as the benchmark. This is the main hall of the building, reconverted from a previously existing Renaissance structure, the old convent of San Lorenzo in Panisperna, which was in turn built on the ruins of a pre-christian roman basilica and of a portion of emperor Nero’s Domus Aurea. The hall consists of two semi-connected rooms, originally the Refectory of the old Convent, that are now used for public events, conferences and graduation ceremonies. This structure can be considered as a model case for similar halls in historical buildings, so that the guidelines deriving from the present study can be extended to other similar environments. The current heating elements are traditional radiators: in our simulations, they have been successively replaced by other elements such as floor and ceiling heating panels and fan coils. Each one of these configurations (the hall and its heating elements) has been modeled and simulated via a commercial CFD code to generate detailed thermal maps and to compute the actual thermal load. Different global “heating chains” were then modeled by coupling solar and hybrid photovoltaic-thermal (PV/T) panels with radiant panels and ground-source heat pumps with fan coils and radiant heating panels. Finally by means of a process simulator software each one of these configurations was analyzed to identify the one that provides the same comfort level with the least exergy use. The procedure also allows to calculate the savings obtained in terms of primary resources.