This paper focuses on optimizing collective self-consumption in an energy community composed of households and premises (stadium) by scheduling loads of electrical appliances owned by the members. The corresponding community remains connected to the public grid, and each member can produce and/or store photovoltaic energy. Furthermore, they can exchange this energy with the public grid or other energy community members. The proposed strategy aims at implementing a Demand Side Management program by taking advantage of the controllable loads' characteristics. A MILP formulation of the problem allows, on the one hand, to give the optimal planning of the operation of the electrical devices. On the other hand, it provides the optimal solutions for managing the storage units and the energy exchanges between community members and the public grid to minimize the energy flows from the public grid to the community over the time horizon. However, this MILP does not allow us to efficiently solve the large instances of the problem. Thus, we develop a column generation-based heuristic to find solutions for large problem instances. Our numerical experiments based on real data collected in the south of France, show that joining an energy community saves money on energy bills and reduces the total energy drawn from the primary grid by at least 15%.