Article ID: | iaor20052136 |
Country: | Netherlands |
Volume: | 157 |
Issue: | 2 |
Start Page Number: | 296 |
End Page Number: | 306 |
Publication Date: | Sep 2004 |
Journal: | European Journal of Operational Research |
Authors: | Mosheiov Gur |
Keywords: | programming: dynamic |
This paper addresses a single-machine scheduling problem with the objective of minimizing a linear combination of total job completion times and total deviation of job completion times from a common due-date. The due-date is assumed to be non-restrictive, i.e., sufficiently large to have no impact on the optimal sequence. When the weights are job-independent, the problem is shown to have a polynomial time solution, and the optimal schedule is fully characterized as a function of the different parameters. When job-dependent weights are assumed, the problem is known to be NP-hard. We introduce a pseudo-polynomial dynamic programming algorithm, indicating that this case is NP-hard in the ordinary sense. The algorithm is shown experimentally to perform extremely well when tested on high-multiplicity instances with up to 1000 jobs.