Order picking policies play an important role in automated storage/retrieval system (AS/RS), and the exact planning can also reduce operational costs of AS/RS. This study integrates the order batching and route planning problems in a man-on-board(MOB) warehousing system. To obtain a better order picking policy, this study searches for a goal solution (or near optimum) by genetic algorithm rather than traditional local searches. An experimental design was made to analyze some design parameters of the genetic algorithm for the order batching and routing planning problem. This results show that the factors of the population size and generations are significant in both solution quality and efficiency in this study.