home » groups » IPROMS 2006 » IPROMS 2006 » Optimisation Techniques

An efficient meta-heuristic for the single machine common due date scheduling problem

Andreas C. Nearchou

A new meta-heuristic for scheduling multiple jobs on a single machine so that they are completed by a common specified date (neither earlier, nor later) is addressed in this paper. Costs are set depending on whether a job finished before (earliness), or after (tardiness) the specified due date. The objective is to minimize the total weighted earliness and tardiness penalized costs from the specified common due date. Minimizing these costs pushes the completion time of each job as close as possible to the due date. Extensive computational experiments over public benchmark problems show the effectiveness of the developed approach. In particular, the proposed meta-heuristic put new improved upper bounds on the majority of the benchmarks test problems.

NOTE: TO READ THIS AND OTHER IPROMS 2006 PAPERS, PLEASE REGISTER FOR THE CONFERENCE.

REGISTRATION IS FREE.

CLICK here TO REGISTER.

‹ Feature selection for SPC chart pattern recognition using fractional factorial experimental designEvolutionary Approach to Measure Production Performance ›
AttachmentSize
PID150210.wmv8.86 MB
PID150210.pdf170.64 KB
a pdf file
Submitted by Nearchou on Sun, 02/07/2006 - 12:32pm.
login or register to post comments |
A warm welcome and a question about bencmarking methodology
Submitted by Ebubekir on Tue, 04/07/2006 - 5:55pm.

Welcome to IPROMS 2006. In this session, we have two new meta-heuristic including this paper. Thanks for your contribution to IPROMS 2006.

I would like to ask you about your bechmarking methodology, as far as I can see, you are not comparing your results with the existing meta-heuristics. Could you explain your bechmarking methodology please?

Thanks

login or register to post comments |
On Behalf of Mr. Nearchou
Submitted by xidias on Fri, 14/07/2006 - 9:54am.

As described in the paper the proposed algorithm was tested over the public benchmarks of the restricted CDDSP, recently proposed by Biskup and Feldmann [4]. These benchmarks include 280 test instances generated with less or more restrictive due dates, ranging from small size instances with 10 jobs to large size instances with 1000 jobs. The upper bounds of the existing optimal solutions for these benchmarks are public available on the web in the web site of the famous OR-Library. More about the benchmarking methodology can be found in section 3 of the paper.

login or register to post comments |
Trying the Bees Algorithm
Submitted by Pham on Fri, 14/07/2006 - 12:59pm.

Hello Professor Nearchou,

It would interesting to try the Bees Algorithm on this problem and compare the results against those produced using DE.

Our Bees Team will be happy to discuss this possibility with you.

Best wishes.

D Pham.

login or register to post comments |

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

User login

Captcha Image: you will need to recognize the text in it.
Please type in the letters/numbers that are shown in the image above.