Advertisement

Springer Nature is making Coronavirus research free. View research | View latest news | Sign up for updates

Auxiliary variables for Bayesian inference in multi-class queueing networks

Abstract

Queueing networks describe complex stochastic systems of both theoretical and practical interest. They provide the means to assess alterations, diagnose poor performance and evaluate robustness across sets of interconnected resources. In the present paper, we focus on the underlying continuous-time Markov chains induced by these networks, and we present a flexible method for drawing parameter inference in multi-class Markovian cases with switching and different service disciplines. The approach is directed towards the inferential problem with missing data, where transition paths of individual tasks among the queues are often unknown. The paper introduces a slice sampling technique with mappings to the measurable space of task transitions between the service stations. This can address time and tractability issues in computational procedures, handle prior system knowledge and overcome common restrictions on service rates across existing inferential frameworks. Finally, the proposed algorithm is validated on synthetic data and applied to a real data set, obtained from a service delivery tasking tool implemented in two university hospitals.

This is a preview of subscription content, log in to check access.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

References

  1. Armero, C., Bayarri, M.J.: Prior assessments for prediction in queues. The Stat. 43(1), 139–153 (1994)

  2. Baskett, F., Chandy, K.M., Muntz, R.R., Palacios, F.G.: Open, closed, and mixed networks of queues with different classes of customers. J. ACM 22(2), 248–260 (1975)

  3. Boyen, X., Koller, D.: Tractable inference for complex stochastic processes. In: Proceedings of the Fourteenth Conference on Uncertainty in Artificial Intelligence, UAI’98, pp 33–42 (1998)

  4. Buzacott, J.A., Shanthikumar, J.G.: Stochastic Models of Manufacturing Systems, vol. 4. Prentice Hall, New Yersey (1993)

  5. Fan, Y., Shelton, CR.: Sampling for approximate inference in continuous time bayesian networks. In: Tenth International Symposium on Artificial Intelligence and Mathematics, ISAIM’08 (2008)

  6. Hobolth, A., Stone, E.A.: Simulation from endpoint-conditioned, continuous-time Markov chains on a finite state space, with applications to molecular evolution. The Ann. Appl. Stat. 3(3), 1204–1231 (2009)

  7. Jensen, A.: Markoff chains as an aid in the study of Markoff processes. Scand. Actuar. J. 36, 87–91 (1953)

  8. Kalli, M., Griffin, J.E., Walker, S.G.: Slice sampling mixture models. Stat. Comput. 21(1), 93–105 (2011)

  9. Kleinrock, L.: Queueing Systems Vol II: Computer Applications. Wiley, New York (1976)

  10. Koole, G., Mandelbaum, A.: Queueing models of call centers: an introduction. Ann. Oper. Res. 113(1), 41–59 (2002)

  11. Lippman, S.A.: Applying a new device in the optimization of exponential queuing systems. Oper. Res. 23(4), 687–710 (1975)

  12. Liu, Z., Wynter, L., Xia, C.H., Zhang, F.: Parameter inference of queueing models for IT systems using end-to-end measurements. Perform. Eval. 63(1), 36–60 (2006)

  13. Neal, R.M.: Slice sampling. The Ann. Stat. 31(3), 705–767 (2003)

  14. Ng, B., Peshkin, L., Pfeffer, A.: Factored particles for scalable monitoring. In: Proceedings of the Eighteenth Conference on Uncertainty in Artificial Intelligence, UAI’02, pp 370–377 (2002)

  15. Nodelman, U., Shelton, CR., Koller, D.: Continuous time bayesian networks. In: Proceedings of the Eighteenth Conference on Uncertainty in Artificial Intelligence, UAI’02, pp 378–387 (2002)

  16. Osorio, C., Bierlaire, M.: An analytic finite capacity queueing network model capturing the propagation of congestion and blocking. Eur. J. Oper. Res. 196(3), 996–1007 (2009)

  17. Perez, I., Brown, M., Pinchin, J., Martindale, S., Sharples, S., Shaw, D., Blakey, J.: Out of hours workload management: Bayesian inference for decision support in secondary care. Artif. Intell. Med. 73, 34–44 (2016)

  18. Rao, V.A., Teh, Y.W.: Fast MCMC sampling for Markov jump processes and extensions. J. Mach. Learn. Res. 14, 3295–3320 (2013)

  19. Spinner, S., Casale, G., Brosig, F., Kounev, S.: Evaluating approaches to resource demand estimation. Perform. Eval. 92, 51–71 (2015)

  20. Sutton, C., Jordan, M.I.: Bayesian inference for queueing networks and modeling of internet services. The Ann. Appl. Stat. 5(1), 254–282 (2011)

  21. Van Gael, J., Saatci, Y., Teh, YW., Ghahramani, Z.: Beam sampling for the infinite hidden markov model. In: Proceedings of the 25th International Conference on Machine Learning, ACM, ICML’08, pp 1088–1095 (2008)

  22. Walker, S.G.: Sampling the dirichlet mixture model with slices. Commun. Stat. Simul. Comput. 36(1), 45–54 (2007)

  23. Wang, W., Casale, G., Sutton, C.: A bayesian approach to parameter inference in queueing networks. ACM Trans. Model. Comput. Simul. 27(1), 2:1–2:26 (2016)

Download references

Acknowledgements

We would like to thank the anonymous reviewers for their valuable remarks and suggestions that have improved the quality of this paper.

Author information

Correspondence to Iker Perez.

Additional information

Work supported by RCUK through the Horizon Digital Economy Research Grants (EP/G065802/1, EP/M000877/1) and The Health Foundation through the Insight 2014 project “Informatics to identify and inform best practice in out of hours secondary care” (7382).

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Perez, I., Hodge, D. & Kypraios, T. Auxiliary variables for Bayesian inference in multi-class queueing networks. Stat Comput 28, 1187–1200 (2018). https://doi.org/10.1007/s11222-017-9787-x

Download citation

Keywords

  • Queueing networks
  • Continuous-time Markov chains
  • Uniformization
  • Markov chain Monte Carlo
  • Slice sampler