This research addresses customer orders scheduling problem on parallel machines, where jobs are dispatched in batches. The objective is to minimize the makespan and maximum lateness, designated as P | batch | Cmax and max P | batch | L , respectively. The problems we considered are NP-hard, therefore both heuristic and optimal algorithms are provided. All heuristics based on simple scheduling rules, for each problem are proposed and their tight worse case bounds are found. Some empirical evaluations for the average-case performance of the heuristics are also performed.