摘要
在分析随机作业调度问题特点的基础上,建立了随机加工时间统计模型及参数估计模型,在参数未知及参数已知的条件下,提出了基于马尔可夫链蒙特卡罗方法的随机加工时间统计技术,并通过吉布斯抽样实现了加工时间的参数估计。通过计算机仿真实验,验证了该方法的可行性及有效性,为随机作业调度提供更符合实际生产的数据支撑。
After analyzing the characteristics of stochastic job-shop scheduling problem, we establish the stochastic processing time statistics model and the parameter estimation model. Under the two conditions that there may exist known or unknown parameters, we propose our stochastic processing time statistics technique based on the Markov chain and the Monte Carlo (MCMC) method. Then we do the Gibbs sampling to estimate the unknown parameters of stochastic processing time. Finally, we carry out the simulation of the technique, and the simulation results verify its feasibility and effectiveness.
出处
《机械科学与技术》
CSCD
北大核心
2007年第12期1574-1577,共4页
Mechanical Science and Technology for Aerospace Engineering
基金
国家863计划项目(2001AA412150
2003AA411110)
教育部博士点基金项目(2004699025)资助
关键词
随机作业调度问题
随机加工时间
马尔可夫链蒙特卡罗方法
吉布斯抽样
stochastic job shop scheduling probelm
stochastic processing time
Markov Chain and Monte Carlo (MCMC) method
Gibbs sampling