中文摘要：

针对当前许多单件小批量生产模式下的企业无法为物料需求计划（MRP）提供完整基础数据支持以及订单交货期不能自动调整的问题,提出了一种单件小批量生产模式下的MRP改进模型.首先研究了MRP的数学计算模型,得出基础数据在数量上和时间上的计算关系;然后使用数据对象多粒度化和订单定位平移时间的方法对MRP模型进行柔性化改进,实现了基础数据的柔性化设置和订单交期的自动调整;最后对MRP改进模型的算法进行了设计、分析与实现.实验结果表明改进后的MRP模型在容错性和自适应性方面有了较大的提高.

英文摘要：

The material requirement planning（ MRP） is based on accurate basic data. But many enterprises in jobbing work circumstance were unable to provide the completed basic data support for MRP. The delivered date of sale order usually needed to be adjusted but it could not be done automatically. To solve these problems,an improved MRP model for jobbing working was proposed. Firstly,the mathematical computing model of MRP was studied to find out the computing relations of basic data on quantity and time. Secondly,the improved MRP model was established with multi-granularity data object and time translation of located orders. In this model,the basic data flexible setting and the orders＇ delivered date automatically adjusting were realized. Lastly,the algorithm of this model was designed,analyzed and implemented. The experimental result shows that the fault tolerance and self-adaption of MRP are greatly improved.