Mathematical Programming

, Volume 98, Issue 1–3, pp 431–444 | Cite as

Asymptotic scheduling

  • Paolo Serafini


We deal with the following scheduling problem: a finite set of jobs is given and each job consists in the execution of an infinite number of tasks. A task is a sequence of operations and each operation requires a specific machine. A machine can process only one operation at a time and preemption is not allowed. Performance measures of the processing system involve fixing a time horizon T, counting the number of tasks completed within T for each job and maximizing a specified function of these numbers to estimate the throughput of the schedule. Whilst computing the throughput for a given T is in general an extremely difficult problem, it is shown in this paper that the limit, as T tends to infinity, of the average throughput (i.e. the throughput divided by T) can be easily computed via Linear Programming under fairly mild conditions. This quantity, which may be called the asymptotic throughput, can be used to assess a bound on performance measures of real systems. Buffers play a crucial role and buffer sizes can be taken care of in assessing the system performance.


scheduling linear programming production throughput 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Applegate, D., Cook, W.: A computational study of job shop scheduling. ORSA J. Comput. 3, (1991)Google Scholar
  2. 2.
    Ball, M.O., Magnanti, T.L., Monma, C.L., Nemhauser, G.L., eds.: Handbooks in Operations Research and Management Science, 7: Network Models, Elsevier Science, Amsterdam, 1995aGoogle Scholar
  3. 3.
    Ball, M.O., Magnanti, T.L., Monma, C.L., Nemhauser, G.L., eds.: Handbooks in Operations Research and Management Science, 8: Network Routing. Elsevier Science, Amsterdam, 1995bGoogle Scholar
  4. 4.
    Garey, M.R., Johnson, D.S.: Computers and intractability: a guide to the theory of NP-completeness. W.H. Freeman and Company, San Francisco, CA, USA, 1979Google Scholar
  5. 5.
    Graves, S.C., Rinnooy Kan, A.H.G., Zipkin, P.H., eds.: Handbooks in Operations Research and Management Science, 4: Logistics of Production and Inventory. Elsevier Science, Amsterdam, 1993Google Scholar
  6. 6.
    Papadimitriou, C., Serafini, P., Yannakakis, M.: Computing the throughput of a network with dedicated lines. Discrete and Applied Mathematics 42, 271–278 (1993)Google Scholar
  7. 7.
    Pochet, Y.: Mathematical programming models and formulations for deterministic production planning problems. Computational combinatorial optimization (Schloß Dagstuhl, 2000) M. Jünger and D. Naddef eds., Springer, Berlin, 2001Google Scholar
  8. 8.
    Silver, E.A., Pyke, D.F., Peterson, R.: Inventory Management and Production Planning and Scheduling. Wiley and Sons, New York, 1998Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Paolo Serafini
    • 1
  1. 1.Italy Dipartimento di Matematica e InformaticaUniversity of UdineUdineItaly

Personalised recommendations