Heat demand in District Heating Systems (DHS) is a highly time-varying quantity because it directly depends on external weather conditions and heat consumption habits. In traditional control strategies of DHS, heat production is contingent on heat demand, thus resulting in peak loads during certain hours of the day. Advanced control strategies relying on Load Shedding (LS) offer an alternative by shifting heat production to off-peak periods. However, ensuring consumers' thermal comfort is of vital importance in this context. Therefore, a careful assessment of all heat gains and losses as well as short-term heat storage capacity due to buildings' thermal inertia including furniture is pivotal for LS. Dynamic Thermal Simulation (DTS) applied to buildings is a powerful tool for the development and assessment of optimal LS control strategies. However, the parameterization of such tools can be rather cumbersome. As a first step towards the development of an LS algorithm, we intend in this communication to present the results of a sensitivity analysis performed to identify the most influencing parameters. Many studies have recognized the important and stochastic impact of occupants on a building's energy balance (e.g. [1]). Most studies considered empty buildings; whereas in [2], the authors emphasized the influence of furniture's thermal mass on heat consumption. Wherefore, we developed a building numerical simulator and we included disturbance signals for internal heat gains and windows opening, as well as additional partition walls representing furniture. Two categories of multi-floors buildings are considered: poorly insulated buildings representing old dwellings, and well-insulated buildings representing recently built ones. Modelica language is used for this end, since it is an object oriented language with a well-suited inherent bottom-up non-causal modeling approach. The MixedAir component of the Buildings library [3] is adopted to model a 4-zones floor. RadiatorEN442_2 is a hydronic radiator model, also found in the Buildings library and used to maintain a set point temperature inside the thermal zones. Dynamic profiles of occupants' behavior and furniture dimensions are varied within certain ranges. Then, the DTS of the model is run over a period of 50 days (Figure 1). A Monte Carlo simulation is used for the sensitivity analysis. Conclusions of this study give recommendations on parameters that should be considered for DHS advanced demand control strategies.