In this paper, we present some of the results of an ongoing research project, which aims at investigating the use of the evolutionary computation paradigm for real world problem solving in an industrial environment. One of the problems targeted in the investigation is that of job sequence optimization for welding robots operating in a shipyard. This is an NP-hard combinatorial optimization problem with constraints. To solve the problem, we propose a hybrid genetic algorithm incorporating domain-specific knowledge. We demonstrate how the method is successful in solving the job sequencing problem. The effectiveness and usefulness of the algorithm is further exemplified by the fact, that it has been implemented in the RoboCopp application program, which is currently used as the task sequence scheduler in a commercially available robot programming environment.