在线咨询
中国工业与应用数学学会会刊
主管:中华人民共和国教育部
主办:西安交通大学
ISSN 1005-3085  CN 61-1269/O1

工程数学学报 ›› 2026, Vol. 43 ›› Issue (1): 103-127.doi: 10.3969/j.issn.1005-3085.2026.01.007cstr: 32411.14.cjem.CN61-1269/O1.2026.01.007

• • 上一篇    下一篇

基于工作量$D$-策略和不中断单重休假的${\rm Geo}^{({\lambda }_1,{\lambda }_2)}/G/1$排队分析

刘琼琳1,  唐应辉1,  魏瑛源2   

  1. 1. 四川师范大学数学科学学院,成都 610068

    2. 河西学院数学与统计学院,张掖 734000

  • 收稿日期:2023-04-19 接受日期:2023-11-08 出版日期:2026-02-15 发布日期:2026-04-15
  • 通讯作者: 唐应辉 E-mail: tangyh@sicnu.edu.cn
  • 基金资助:
    国家自然科学基金 (71571127);四川师范大学学科建设专项 (XKZX2021-04);河西学院校长基金创新团队项目 (CXTD2022013).

Analysis of ${\rm Geo}^{({{\lambda }_{1}},{{\lambda }_{2}})}/G/1$ Queue with Workload $D$-policy and Uninterrupted Single Vacation

LIU Qionglin1,  TANG Yinghui1,  WEI Yingyuan2   

  1. 1. School of Mathematical Sciences, Sichuan Normal University, Chengdu 610068
    2. School of Mathematics & Statistics, Hexi University, Zhangye 734000
  • Received:2023-04-19 Accepted:2023-11-08 Online:2026-02-15 Published:2026-04-15
  • Contact: Y. Tang. E-mail address: tangyh@sicnu.edu.cn
  • Supported by:
    The National Natural Science Foundation of China (71571127); the Specialized Project for Subject Construction of Sichuan Normal University (XKZX2021-04); the Innovation Team Fund of the President of Hexi University (CXTD2022013).

摘要:

研究了在服务员工作量$D$-策略控制下,不中断单重休假和可变到达率的${\rm Geo}^{(\lambda_1,\lambda_2)}/G/1$排队的瞬态与稳态性能。使用全概率分解技术和更新过程理论,详细分析了在任意初始状态下队长的瞬态概率分布,得到了在任意时刻${{n}^{+}}$瞬态队长分布的$z$-变换表达式。在瞬态分析的基础上,通过一些代数运算得到了稳态队长分布的递推表达式,同时推导了系统其他重要的排队性能指标。进一步得到了系统在不同时刻点$n$、${{n}^{-}}$处队长的稳态概率分布的表达式。最后,在建立的费用模型下用数值例子讨论了系统的最小成本问题,并确定了使得系统在长期单位时间内期望 成本最小的最优控制策略。

关键词: 服务员工作量$D$-策略, 不中断单重休假, 可变到达率, 全概率分解技术, 最优控制策略

Abstract:

This paper studies the transient and steady-state performance of the ${\rm Geo}^{({{\lambda }_{1}},{{\lambda }_{2}})}/G/1$ queue with uninterrupted single vacation and variable arrival rates under the control of server's workload $D$-policy. Employing the total probability decomposition technique and the renewal process theory, we present a detailed analysis of the transient probability distribution of the queue length under any initial state, and obtain the expressions of the $z$-transformation of the transient queue length distribution at any epoch ${{n}^{+}}$. Based on the transient analysis, the recursive expressions of the steady-state queue length distribution are derived by using some algebraic operation. Meanwhile, some other important queueing performance measures of the system are obtained. Furthermore, the expressions of the steady-state probability distribution of the queue length at different epochs $n$ and ${{n}^{-}}$ are obtained. Finally, the minimum cost problem of the system is discussed with numerical examples under the established cost model, and the optimal control policies that minimize the long-run expected cost per unit time are determined.

Key words: server's workload $D$-policy, uninterrupted single vacation, variable arrival rates, total probability decomposition technique, optimal control policies

中图分类号: