| 备件需求过程的MAP数据拟合算法研究 |
| |
| 引用本文： | 陈童,狄鹏,郭波.备件需求过程的MAP数据拟合算法研究[J].海军工程大学学报,2009,21(1). |
| |
| 作者姓名： | 陈童 狄鹏 郭波 |
| |
| 作者单位： | 1. 国防科技大学,信息系统与管理学院,长沙,410073
2. 海军工程大学,管理工程系,武汉,430033 |
| |
| 摘 要： | 为了将备件需求过程表示为马尔可夫到达过程(Markovian arrival process, MAP)形式,设计了备件需求到达流的MAP拟合算法.首先,采用EM算法将备件需求到达间隔时间拟合成Hyper-Erlang分布形式,然后利用MAP性质和Bayes公式推导出生成元矩阵的计算公式;随后设计了一个完整的数据拟合流程,并通过实例对算法的效果和效率与已有研究进行了对比.结果表明,该算法在确保拟合效果的同时,能够有效提升拟合效率.
|
| 关 键 词： | 备件 需求过程 马尔可夫到达过程 数据拟合 |
An algorithm for fitting MAP to spare parts demand process |
| |
| CHEN Tong,DI Peng,GUO Bo.An algorithm for fitting MAP to spare parts demand process[J].Journal of Naval University of Engineering,2009,21(1). |
| |
| Authors: | CHEN Tong DI Peng GUO Bo |
| |
| Institution: | 1.College of Information System and Management;National Univ.of Defense Technology;Changsha 410073;China;2.Dept.of Management Science;Naval Univ.of Engineering;Wuhan 430033;China |
| |
| Abstract: | A two-step algorithm was proposed to fit the parameters of Markovian arrival pro-cess(MAP) for the spare parts demand arrival stream.Firstly,an EM algorithm was used to fit the inter-arrival time trace into Hyper-Erlang distribution.Then the infinitesimal generator matrix was constructed by using the properties of MAP and Bayes formula.In the end,a complete fitting procedure was designed.The numerical examples show that this algorithm can raise the efficiency through the contrast with the other researches. |
| |
| Keywords: | spare parts demand process Markovian arrival process data fitting |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
