Optimal server allocation in closed finite queueing networks

  • J. MacGregor Smith
  • Ryan Barnes


Many topological network design problems in manufacturing and service systems can be modelled with closed queueing networks, finite buffers, and multiple-servers. Because of their integrality, it is difficult to predict the performance of these problems, let alone optimize their parameters. This paper presents a new queue decomposition approach for the modelling of these finite buffer closed queueing networks with multiple servers along with an integrated approach to the optimization of the allocation of the servers in the network topology. The performance and optimization algorithms are described and a number of experiments for series, merge, and split topologies together with the integration of material handling and layout systems are carried out.


Closed Finite networks \(M/M/c/K\) and \(M/G/c/c\) queues 



We are indebted to Klaus Schittkowsi and his colleagues for letting us use their sequential quadratic nonlinear programming routine along with the branch-and-bound code and embedding it in our queue decomposition methodology. Overall, it performed very well. We are also indebted to the referees for their cogent and insightful comments on the drafts of the paper.


  1. Akyildiz IF (1988) On the exact and approximate throughput analysis of closed queueing networks with blocking. IEEE Trans Softw Eng 14(4):62–71CrossRefGoogle Scholar
  2. Bolch G, Greiner S, de Meer H, Trivedi KS (1998) Queueing networks and Markov chains. Wiley, New YorkCrossRefMATHGoogle Scholar
  3. Buzzacott J, Shantikumar JG (1993) Stochastic models of manufacturing systems. Prentice-Hall, Englewood CliffsGoogle Scholar
  4. Buzen JP (1973) Computational algorithms for closed queueing networks with exponential servers. Commun ACM 16(9):527–531CrossRefMATHMathSciNetGoogle Scholar
  5. Dallery Y, Stecke KE (1990) On the optimal allocation of servers and workloads in closed queueing networks. Oper Res 38(4):694–703CrossRefMATHMathSciNetGoogle Scholar
  6. Exler O, Schittkowski K (2007) A trust region SQP algorithm for mixed-integer nonlinear programming. Optim Lett 1:269–280CrossRefMATHMathSciNetGoogle Scholar
  7. Exler O, Lehmann T, Schittkowski K (2012) MISQP: A Fortran subroutine of a trust region SQP algorithm for mixed-integer nonlinear programming-users guide. Department of Computer Science, University of Bayreuth, GermanyGoogle Scholar
  8. Exler O, Lehmann T, Schittkowski K (2013) A comparative study of SQP-type algorithms for nonlinear and nonconvex mixed-integer optimization. Math Progr Comput 4:383–412CrossRefMathSciNetGoogle Scholar
  9. Hillier FS, So KC (1989) The assignment of extra servers to stations in tandem queueing systems with smll or no buffers. Perform Eval 10:219–231CrossRefGoogle Scholar
  10. Hillier FS, So KC (1995) On the optimal design of tandem queueing systems with finite buffers. Questa 21(3–4):245–266MATHMathSciNetGoogle Scholar
  11. Hillier FS, So KC (1996) On the simultaneous optimization of server and work allocations in production line systems with variable processing timesations. Oper Res 44(3):435–443CrossRefMATHGoogle Scholar
  12. Kelton D, Sadowski R, Swets N (2010) Simulation with Arena, 5th edn. McGraw-Hill, New YorkGoogle Scholar
  13. Kerbache L, Smith JM (1988) Asymptotic behavior of the expansion method for open finite queueing networks. Comput Oper Res 15(2):157–169CrossRefMATHGoogle Scholar
  14. Kobayashi H, Mark BL (1994) On queueing networks and loss networks. In: Proceedings of the 28th annual conference on information sciences and systems. Princeton, New Jersey, pp 794–799Google Scholar
  15. Kobayashi H, Mark BL (2002) Generalized loss models and queueing loss models. ITOR 9:97–112MATHGoogle Scholar
  16. Lee HF, Srinivasan MM, Yano C (1991) Characteristics of optimal workload allocation for closed queueing networks. Perform Eval 12:255–268CrossRefMATHMathSciNetGoogle Scholar
  17. Magazine M, Stecke K (1996) Throughput for production lines with serial workstations and parallel service facilities. Perform eval 25:211–232CrossRefMATHGoogle Scholar
  18. Onvural R (1990) Survey of closed queueing networks with blocking. ACM Comput Surv 22(2):83–121CrossRefGoogle Scholar
  19. Reiser M, Lavenberg S (1980) Mean value analysis of closed multi-chain queueing networks. JACM 27(2):313–322CrossRefMATHMathSciNetGoogle Scholar
  20. Shantikumar JG, Yao DD (1987) Optimal server allocation in a system of multi-server stations. Manag Sci 33(9):1173–1180CrossRefGoogle Scholar
  21. Smith JM (2014) Queue decomposition and finite closed queueing network models. Working paper. Department of Mechanical and Industrial Engineering, University of Massachusetts, Amherst, MAGoogle Scholar
  22. Smith JM, Cruz FRB, van Woensel T (2010) Optimal server allocation in general, finite, multi-server queueing networks. Appl Stoch Models Bus Ind 26:705–736CrossRefMATHMathSciNetGoogle Scholar
  23. Spinellis D, Papadopoulos CT, Smith JM (2000) Large production line optimization using simulated annealing. Int J Prod Res 38:509–541CrossRefMATHGoogle Scholar
  24. Suri R, Sahu S, Vernon M (2007) Approximate mean value analysis for closed queueing networks with multiple-server stations. In: Bayraksan G, Lin W, Wysk R (eds) Proceedings of the 2007 IERCGoogle Scholar
  25. Van Woensel T, Andriasyah R, Cruz FRB, Smith JM, Kerbache L (2010) Buffer and server allocation in general multi-server networks. ITOR 17:257–286MATHGoogle Scholar
  26. Yarmand MH (2013) Optimal allocation in zero-buffer tandem queues. Open access dissertations and theses. Paper no. 7690Google Scholar
  27. Yuzukirmizi M (2005) Closed finite queueing networks with multiple servers and multiple customer types. Ph.D. Dissertation. Department of Industrial Engineering and Operations Research at the University of Massachusetts, Amherst Campus, March 2005Google Scholar

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  1. 1.Department of Mechanical and Industrial EngineeringUniversity of Massachusetts AmherstAmherstUSA

Personalised recommendations