模具生产过程中充满了不确定性,尤其是返修的工序和次数不确定,使得准确估计和控制模具的生产周期成为一个世界性难题。为了解决这一问题,分别采用考虑返修的串联和混联制造排队系统来描述模具制造过程,在假定订单按先到先服务(First come first served,FCFS)规则进行加工且正常加工和返修时间均服从负指数分布的情况下,分析制造排队系统基于事件的状态转移过程和事件间隔时间,提出了返修强度的概念并将事件间隔时间近似为负指数分布,在此基础上结合独立伽马随机变量之和推导出模具剩余工期的近似数学分布模型。设计计算实例,将不同返修强度和初始状态下的数值计算结果与仿真结果进行对比分析,验证了剩余工期近似模型在不同返修强度环境下均具有良好可靠性。
Mold production process is fraught with uncertainty, especially the uncertainty of process and number of reworks which has made accurate estimates and control mold production cycle to become a worldwide problem. To address this issue, tandem and hybrid manufacturing queuing system considering reworks is used to presents the mold manufacturing process, respectively. Under the assumption that the orders are processing under the first come first served principal and both the normal processing and rework time are exponential distributed, the event-based state transition process of the manufacturing queuing system has been analyzed, meanwhile, the concept of the rework strength is proposed and the event interval is approximately exponential distribution. On that basis, the mold remaining duration approximate mathematical distribution model is deducted by combining the sum of independent gamma random variables. A comparative analysis of the numerical results and simulation results under different rework strength and initial states verified the reliability of proposed approximate model of remaining duration.