Additive manufacturing (AM) is a promising technology for the rapid prototyping and production of highly customized products. The scheduling of AM machines has an essential role in increasing profitability and has recently received a great deal of attention. This paper investigates the scheduling of batch processing of parallel 3d-printing machines to minimize the total weighted tardiness. Accordingly, a mathematical model is proposed to formulate the problem considering the sequence-dependent setup time and incompatible job families, where jobs of different families are processed with different materials and desired quality. Due to the high complexity of the problem, an efficient matheuristic algorithm is presented based on the hybridization of a genetic algorithm and a local search method based on mixed integer programming (MIP). Computational results show that the proposed approach is efficient and promising to solve the problem.