研究利用前向神经网络 (也称多层感知机 )安排单件小批量订单生产类型企业的生产作业计划 ,针对数据难于组织、反传算法易陷入局部极小点和输出结果存在偏差的问题 ,提出了以下的解决方法 :基于位混合编码的样本组织方式 ;权值和迁跃点记忆队列的网络训练方法和结果校正方法 .仿真结果证明了解决方法的有效性和基于神经网络研究生产作业计划的可行性 . In order to solve the problem that the data is hard to be organized and the anti-pass algorithm is easy to fall into the local minimum and the output result is biased, this paper studies how to arrange production planning of single-lot and small-lot order production enterprises by using forward neural network (also called multi- The following solutions are proposed: sample organization based on bit hybrid coding; network training method and result correction method of weight and transition point memory queue.The simulation results prove the effectiveness of the solution and neural network based on the study of production planning Feasibility.