An innovative maintenance policy is proposed in this paper. This policy can involve spare parts ordering, production quality, and buffer inventory for an efficient production system. In the system, certain batches are required to be produced, and when each batch is finished, a determination is made whether maintenance is needed. The machine state deteriorates with the number of completed batches, and it can be improved by performing maintenance. Two types of maintenance activity, replacement and imperfect maintenance, can be selectively chosen to minimize cost. The defect rate of each batch is related to the number of completed production batches. An innovative concept, defined as the “virtual number” of completed production batches, is used to establish a link between maintenance and defect rate. Monte Carlo simulation and enumerative search is then used to determine cost-effective spare parts ordering and maintenance policies to minimize the cost for the production cycle. Finally, numerical examples are presented to demonstrate the model and to conduct sensitivity analysis. We find that in situations with a high buffer inventory costs, spare parts should be ordered late. When increasing the buffer inventory cost, more replacements should be performed compared to imperfect maintenance. Also, the buffer inventory cost rate and replacement duration time effect the rate of defective products significantly. These two parameters should be kept small, and controlled, if a very low defect rate is needed.

本文提出了一种创新的维护策略。该政策可能涉及高效生产系统的备件订购、生产质量和缓冲库存。在系统中，需要生产一定的批次，当每个批次生产完成后，再判断是否需要维修。机器状态随着完成批次的数量而恶化，可以通过进行维护来改善。可以选择性地选择两种类型的维护活动，更换和不完善维护，以最大限度地降低成本。每批次的不良率与完成的生产批次数有关。定义为已完成生产批次的“虚拟数量”的创新概念用于在维护和缺陷率之间建立联系。然后使用蒙特卡罗模拟和枚举搜索来确定具有成本效益的备件订购和维护政策，以最大限度地降低生产周期的成本。最后，给出了数值例子来证明模型并进行敏感性分析。我们发现，在缓冲库存成本较高的情况下，应延迟订购备件。当增加缓冲库存成本时，与不完善的维护相比，应该进行更多的更换。此外，缓冲库存成本率和更换持续时间显着影响不良品率。如果需要非常低的缺陷率，这两个参数应该保持较小并受到控制。给出了数值例子来演示模型并进行敏感性分析。我们发现，在缓冲库存成本较高的情况下，应延迟订购备件。当增加缓冲库存成本时，与不完善的维护相比，应该进行更多的更换。此外，缓冲库存成本率和更换持续时间显着影响不良品率。如果需要非常低的缺陷率，这两个参数应该保持较小并受到控制。给出了数值例子来演示模型并进行敏感性分析。我们发现，在缓冲库存成本较高的情况下，应延迟订购备件。当增加缓冲库存成本时，与不完善的维护相比，应该进行更多的更换。此外，缓冲库存成本率和更换持续时间显着影响不良品率。如果需要非常低的缺陷率，这两个参数应该保持较小并受到控制。缓冲库存成本率和更换持续时间对次品率影响显着。如果需要非常低的缺陷率，这两个参数应该保持较小并受到控制。缓冲库存成本率和更换持续时间对次品率影响显着。如果需要非常低的缺陷率，这两个参数应该保持较小并受到控制。