In this paper, the problem of batch scheduling in parallel machines environment with the objective of minimizing make span (Cmax) is addressed. The main contribution of this research is the stochastic nature of the processing times of jobs and release times for better depiction of the real world. It has been proved that the problem is NP-hard. Therefore, we apply heuristic approaches to solve this problem. The provided problem includes two stages of decision making. In the first stage, the jobs are classified into batches and in the next stage; these batches should be assigned to parallel machines. Two and three heuristic methods are used for producing batches and sequencing batches, respectively. 10, 000 test problems are randomly generated due to stochastic nature of processing times and release dates. Using the results of simulating test problems, six combinations of heuristic methods are compared. The results show that applying MBF heuristic method in the first stage and ERT-LPT method in the second stage provide better and efficient solutions.