A Marginal Productivity Index Rule for Scheduling Multiclass Queues with Setups

  • José Niño-Mora
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5425)


This paper addresses the problem of designing a tractable scheduling rule for a multiclass M/G/1 queue incurring class-dependent linear holding costs and setup costs, as well as class-dependent generally distributed setup times, which performs well relative to the discounted or average cost objective. We introduce a new dynamic scheduling rule based on priority indices which emerges from deployment of a systematic methodology for obtaining marginal productivity index policies in the framework of restless bandit models, introduced by Whittle (1988) and developed by the author over the last decade. For each class, two indices are defined: an active and a passive index, depending on whether the class is or is not set up, which are functions of the class state (number in system). The index rule prescribes to engage at each time a class of highest index: it thus dynamically indicates both when to leave the class being currently served, and which class to serve next. The paper (i) formulates the problem as a semi-Markov multiarmed restless bandit problem; (ii) introduces the required extensions to previous indexation theory; and (iii) gives closed index formulae for the average criterion.


Stochastic scheduling optimal service control of queues multiclass queues setup times setup costs polling systems index policies marginal productivity index queues with hysteresis 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Eisenberg, M.: The polling system with a stopping server. Queueing Syst. 18, 387–431 (1994)MathSciNetCrossRefzbMATHGoogle Scholar
  2. 2.
    Takagi, H.: Analysis of Polling Systems. MIT Press, Cambridge (1986)Google Scholar
  3. 3.
    Levy, H., Sidi, M.: Polling systems: Applications, modelling and optimization. IEEE Trans. Comm. 38, 1750–1760 (1990)CrossRefGoogle Scholar
  4. 4.
    Hofri, M., Ross, K.W.: On the optimal control of two queues with server setup times and its analysis. SIAM J. Comput. 16, 399–420 (1987)MathSciNetCrossRefzbMATHGoogle Scholar
  5. 5.
    Boxma, O.J., Levy, H., Weststrate, J.A.: Efficient visit frequencies for polling tables: Minimization of waiting cost. Queueing Syst. 9, 133–162 (1991)CrossRefzbMATHGoogle Scholar
  6. 6.
    Boxma, O.J., Levy, H., Weststrate, J.A.: Efficient visit frequencies for polling systems. Performance Evaluation 18, 103–123 (1993)CrossRefzbMATHGoogle Scholar
  7. 7.
    Duenyas, I., Van Oyen, M.: Heuristic scheduling of parallel heterogeneous queues with setups. Management Sci. 42, 814–829 (1996)CrossRefzbMATHGoogle Scholar
  8. 8.
    Duenyas, I., Van Oyen, M.: Stochastic scheduling of parallel queues with setup costs. Queueing Syst. 19, 421–444 (1995)MathSciNetCrossRefzbMATHGoogle Scholar
  9. 9.
    Reiman, M.I., Wein, L.M.: Dynamic scheduling of a two-class queue with setups. Oper. Res. 46, 532–547 (1998)MathSciNetCrossRefzbMATHGoogle Scholar
  10. 10.
    Olsen, T.L.: A practical scheduling method for multiclass production systems with setups. Management Sci. 45, 116–130 (1999)CrossRefzbMATHGoogle Scholar
  11. 11.
    Lan, W.M., Olsen, T.L.: Multiproduct systems with both setup times and costs: fluid bounds and schedules. Oper. Res. 54, 505–522 (2006)MathSciNetCrossRefzbMATHGoogle Scholar
  12. 12.
    Bertsimas, D., Niño-Mora, J.: Optimization of multiclass queueing networks with changeover times via the achievable region approach: Part I, the single-station case. Math. Oper. Res. 24, 306–330 (1999)MathSciNetCrossRefzbMATHGoogle Scholar
  13. 13.
    Whittle, P.: Restless bandits: Activity allocation in a changing world. In: Gani, J. (ed.) A Celebration of Applied Probability; J. Appl. Probab. Applied Probability Trust 25A (spec. vol.), 287–298, Sheffield, UK (1988)Google Scholar
  14. 14.
    Gittins, J.C., Jones, D.M.: A dynamic allocation index for the sequential design of experiments. In: Gani, J., Sarkadi, K., Vincze, I. (eds.) Progress in Statistics (European Meeting of Statisticians, Budapest, 1972), pp. 241–266. North-Holland, Amsterdam (1974)Google Scholar
  15. 15.
    Niño-Mora, J.: Restless bandits, partial conservation laws and indexability. Adv. Appl. Probab. 33, 76–98 (2001)MathSciNetCrossRefzbMATHGoogle Scholar
  16. 16.
    Niño-Mora, J.: Dynamic allocation indices for restless projects and queueing admission control: a polyhedral approach. Math. Program. 93, 361–413 (2002)MathSciNetCrossRefzbMATHGoogle Scholar
  17. 17.
    Niño-Mora, J.: Restless bandit marginal productivity indices, diminishing returns and optimal control of make-to-order/make-to-stock M/G/1 queues. Math. Oper. Res. 31, 50–84 (2006)MathSciNetCrossRefzbMATHGoogle Scholar
  18. 18.
    Niño-Mora, J.: Marginal productivity index policies for scheduling a multiclass delay-/loss-sensitive queue. Queueing Syst. 54, 281–312 (2006)MathSciNetCrossRefzbMATHGoogle Scholar
  19. 19.
    Niño-Mora, J.: A faster index algorithm and a computational study for bandits with switching costs. INFORMS J. Comput. 20, 255–269 (2008)MathSciNetCrossRefzbMATHGoogle Scholar
  20. 20.
    Niño-Mora, J.: Dynamic priority allocation via restless bandit marginal productivity indices. TOP 15, 161–198 (2007)MathSciNetCrossRefzbMATHGoogle Scholar
  21. 21.
    Niño-Mora, J.: Computing an index policy for bandits with switching penalties. In: ValueTools 2007: Proceedings of the Second International Conference on Performance Evaluation Methodologies and Tools (Nantes, France). ACM International Conference Proceedinsg Series, ICST, Brussels, Belgium (2007)Google Scholar
  22. 22.
    Niño-Mora, J.: Marginal productivity index policies for scheduling restless bandits with switching penalties. In: Albers, S., Möhring, R.H., Pflug, G.C., Schultz, R. (eds.) Algorithms for Optimization with Incomplete Information. Number 05031 in Dagstuhl Seminar Proceedings (2005)Google Scholar
  23. 23.
    Yadin, M., Naor, P.: Queueing systems with a removable service station. Oper. Res. 14, 393–405 (1963)CrossRefGoogle Scholar
  24. 24.
    Bell, C.E.: Characterization and computation of optimal policies for operating an M/G/1 queuing system with removable server. Oper. Res. 19, 208–218 (1971)CrossRefzbMATHGoogle Scholar
  25. 25.
    Bell, C.E.: Optimal operation of an M/G/1 priority queue with removable server. Oper. Res. 21, 1281–1290 (1973)MathSciNetCrossRefzbMATHGoogle Scholar
  26. 26.
    Heyman, D.: Optimal operating policies for M/G/1 queuing systems. Oper. Res. 16, 362–382 (1968)CrossRefzbMATHGoogle Scholar
  27. 27.
    Sobel, M.J.: Optimal average-cost policy for a queue with start-up and shut-down costs. Oper. Res. 17, 145–162 (1969)MathSciNetCrossRefzbMATHGoogle Scholar
  28. 28.
    Borthakur, A., Medhi, J., Gohain, R.: Poisson input queueing system with startup time and under control-operating policy. Comput. Oper. Res. 14, 33–40 (1987)MathSciNetCrossRefzbMATHGoogle Scholar
  29. 29.
    Liu, Z., Nain, P., Towsley, D.: On optimal polling policies. Queueing Syst. 11, 59–83 (1992)MathSciNetCrossRefzbMATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • José Niño-Mora
    • 1
  1. 1.Department of StatisticsUniversidad Carlos III de MadridLeganés (Madrid)Spain

Personalised recommendations