Currently, the idea of designing collaborative smart buildings is becoming a real challenge. Indeed, the emergence of smart grids and the growing use of smart household appliances is the main driving force behind the use of demand side management. This would require developing an efficient control strategy of in-home appliances; especially with the presence of increasing ownership of distributed generations. The main challenge is to provide the capability to control electrical appliances in the framework of demand side management (DSM). Therefore, in this project a novel approach for DSM-based collaborative smart building has been proposed. The approach is based on a dynamic algorithm which strives to resolve two successive multi-objective optimization problems. The first gives activation scheduling towards reshaping the power consumption of non-flexible Deferrable Loads and the second presents the power profiles of flexible Deferrable Loads. The results show the validity of this approach to deal with tracking renewable generation, or generally with any required shaping of the demand curve.