The main purpose of this paper is to develop a hybrid genetic algorithm in order to determine the lot sizes and their production scheduling in a single machine manufacturing system for multi-item orders, the objective function minimizes the sum of holding costs, tardy costs and setup costs. The problem considers a set of orders to be processed each one with its own due date. Each order must be delivered complete. In the scheduling are considered sequence dependent setup times. The proposed hybrid genetic algorithm has embedded a heuristic that is used to calculate its fitness function. The heuristic method presents a modification on the optimal timming algorithm in which are involved sequence dependent set up times. A design of experiments is developed in order to assess the algorithm performance, which is also tested using random-generated data and results are compared with those generated by an exact method. The results show that the algorithm achieves a good performance in both solution quality and time especially for large instances.