Japan Long Term Visit: Queueing Theory, with Applications to Manufacturing Systems and Communications Networks

日本长期访问：排队理论及其在制造系统和通信网络中的应用

• 批准号: 9107700
• 负责人: Ronald Wolff
• 金额: $ 7.81万
• 依托单位: University of California-Berkeley
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 1991
• 资助国家: 美国
• 起止时间: 1991-09-01 至 1993-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=9107700&HistoricalAwards=false
• 关键词: Japan; Long; Term; Visit; Queueing

This award will support a long-term visit by Professor Ronald W. Wolff, Department of Industrial Engineering and Operations Research, Engineering Systems Research Center, University of California, Berkeley, to Japan for a cooperative research project with Professor Masakiyo Miyazawa at the Science University of Tokyo and Professor Genji Yamazaki at the Tokyo Metropolitan Institute of Technology. The researchers will investigate several topics in queueing theory that have applications to model manufacturing systems and communications networks as follows: 1) Bounds and Approximations for Queues and Queueing Networks. Generally performance measure for queueing networks cannot be determined exactly, but must be approximated. They will investigate the errors in these approximations in several ways including light-traffic analysis. For tandem queues they will investigate bounds on the difference in performance measures for different arrangements of the order of the servers. They will also look at some unsolved problems about bounds and approximations for simple single-and multi-server queues. 2) Conditions for Finite Delay Moments. The formal derivation of approximations and bounds often involve the requirement that certain terms either cancel or are small. This in turn requires that certain delay moments be finite. Recent progress has been made in this area and the researchers believe they can contribute further knowledge to the problem. 3) Fundamental Properties of Queues. They will investigate applications of attained waiting time to queues, where one of the objectives is to find new bounds on performance measures. Finally, they will investigate simplifications to existing theory that may result from the merger of different approaches, such as the application of sample-path methods to the marked point processes. There is a large amount of research currently taking place on queueing theory in Japan. Thus it is an excellent opportunity for Professor Wolff, who is one of the outstanding queueing theorists in the world, to collaborate with Professors Miyazawa and Yamazaki who are also world renown in the field. Each researcher brings a unique capability to the effort yet each is also experienced in the work of the others. It is expected that this collaboration should provide some exciting results as well as continued cooperation between the researchers for years to come.

该奖项将支持罗纳德教授的长期访问 W.沃尔夫，工业工程与运营系 研究，工程系统研究中心，大学 加州伯克利，日本进行合作研究 与Masakiyo Miyazawa教授在科学 东京大学教授山崎亘二在东京 大都会理工学院。 研究人员将 研究了几个主题的理论， 应用于制造系统和通信的建模 网络如下：1）边界和近似的时间 和网络公司。 一般而言， 网络不能被精确地确定，但必须 近似。 他们将调查这些错误 在几个方面的近似，包括轻交通 分析. 对于串联队列，他们将研究 不同的绩效指标之间的差异 服务器的顺序。 他们也会看 在一些未解决的问题的界限和近似， 简单的单服务器和多服务器队列。 2)条件 有限延迟时刻。 近似的形式推导 和界限往往涉及到某些条款的要求， 要么取消，要么很小。 这反过来又要求， 某些延迟时刻是有限的。 最近的进展是 研究人员认为， 为这个问题提供更多的知识。 3)基本 的属性。 他们将研究 达到排队等待时间，其中一个目标 是找到新的绩效衡量标准。 最后， 他们将研究对现有理论的简化， 可能是由于合并了不同的方法，例如 样轨法在标志点上的应用 流程. 目前有大量的研究正在进行， 在日本的继承理论。 因此这是一个极好的机会 沃尔夫教授，他是一个杰出的 世界上的理论家，与教授合作， 宫泽和山崎也是世界知名的 在外地 每一位研究人员都有独特的能力， 的努力，但每个人也有经验的工作， 他人 预计这一合作将 提供了一些令人兴奋的成果， 研究人员之间的合作