A sensitivity analysis to assess the completion time deviation for multi-purpose machines facing demand uncertainty
- 77 Downloads
This paper addresses multi-purpose machine configuration in an uncertain context through sensitivity analysis. The so-called configuration is the machine’s ability to process products, and the uncertain context is modelled as a demand variation affecting the forecast demand. Given a configuration, this work aims at assessing the completion time deviation when the workshop demand is subject to perturbation. Such quantitative information can be used in a robustness approach for selecting the most appropriate configuration. To do so, the configuration impact on the completion time value that can be reached by solving the attached scheduling problem is first investigated. Then, the completion time deviation is written as a piecewise linear function of the magnitude of demand variation. The proposed approach, which is based on the solution of a set of linear programs, is illustrated through a detailed example. It is shown to be polynomial, and fast enough for addressing real-world instances. Finally, how to compare two configurations on the basis of completion time deviation in an uncertain context is demonstrated.
KeywordsSensitivity analysis Multi-purpose machines Linear programming Configuration Demand uncertainty
Unable to display preview. Download preview PDF.
- Aubry, A., Rossi, A., & Jacomino, M. (2009). A generic off-line approach for dealing with uncertainty in optimisation. In Preprints of the 13th IFAC symposium on information control problems in manufacturing, Moscow, Russia, 3–5 June (pp. 1464–1469). Google Scholar
- Espinouse, M.-L., Jacomino, M., & Rossi, A. (2008). The robustness of multi-purpose machines workshop configuration. In Flexibility and robustness in scheduling. London: ISTE Ltd, Wiley. Google Scholar
- Kouvelis, P., & Yu, G. (1997). Robust discrete optimization and its applications. Dordrecht: Kluwer Academic. Google Scholar
- Lawler, E., & Labetoulle, J. (1978). On preemptive scheduling of unrelated parallel processors by linear programming. Journal of the Association for Computing Machinery, 25, 612–619. Google Scholar
- Rossi, A. (2003). Ordonnancement en milieu incertain, mise en œuvre d’une démarche robuste. Ph.D. thesis, Institut National Polytechnique de Grenoble, France. Google Scholar