This paper incorporates the maintenance time schedule optimization into the maintenance planning problem for building energy efficiency retrofitting project. An impulsive switched system model is employed to describe the aggregate performance and population dynamics of the retrofitted items. The maintenances are considered as impulsive effects that switch the items to different working states. The control objective is to optimize the combination of maintenance instants and intensities to maximize the overall energy savings and financial payback of the retrofitting project against deteriorations. The feasibility and boundedness of the optimal control problem is discussed. A multi-scale Differential Evolution algorithm is proposed as a numerical solver, which optimizes the maintenance time schedule and intensities simultaneously. Simulation results illustrate the effectiveness of the proposed approach.