A metaheuristic is generally a procedure designed to find a good solution to a difficult optimization problem. Known optimization search metaheuristics heavily rely on parameters, which are usually introduced so that the metaheuristic follows some supposedly related to the optimization problem natural process (simulated annealing, swarm optimization, genetic algorithms). Adjusting the parameters so that the metaheuristic performs successfully in the problem at hand could be quite tricky and time consuming task which often requires intimate knowledge of the problem and a lot of experimenting to achieve the needed level of performance. In this article I present a metaheuristic with parameters depending only on the problem at hand, which virtually eliminates the preliminary work on adjusting the parameters. Moreover, the parameters are frequently updated during the process, based on the increasing amount of information about the solution space collected during the run. The metaheuristic has been successfully applied in several different searches for discrete objects such as designs, packings, coverings and partitions.
This is a preview of subscription content, access via your institution.
Buy single article
Instant access to the full article PDF.
Tax calculation will be finalised during checkout.
Abel J, Assaf A, Bennett F, Bluskov I, Greig M (2006) Pair covering designs with block size 5. Discret Math 307:1776–1791
Abel J, Assaf A, Bluskov I, Greig M, Shalabi N (2010) New results on GDDs, covering, packing and directable designs with block size 5 with higher index. J Comb Des 18:337–368
Abel J, Bluskov I, de Heer J, Greig M (2007) Pair covering and other designs with block size 6. J Comb Des 15:511–533
Abel J, Bluskov I, Greig M (2001) Balanced incomplete block designs with block size 8. J Comb Des 9:233–268
Abel J, Bluskov I, Greig M (2002) Balanced incomplete block designs with block size 9 and \(\lambda = 2,4\) and 8. Codes, Des Cryptogr 26:33–59
Abel J, Bluskov I, Greig M (2004a) Balanced incomplete block designs with block size 9: Part II. Discret Math 279:5–32
Abel J, Bluskov I, Greig M (2004b) Balanced incomplete block designs with block size 9: Part III. Australas J Comb 30:57–73
Bertolo R, Bluskov I, Hammalainen H (2004) Upper bounds on the general covering number \(C(v, k, t, m)\). J Comb Des 12:362–380
Bluskov I (in press) Some new upper bounds on the size of constant weight codes
Bluskov I (2007) On the covering numbers \(C_2(v, k, t)\). J Comb Math Comb Comput 63:17–32
Bluskov I, Greig M (2006) Pair covering designs with block size 5 with higher index—the case of \(v\) even. J Comb Math Comb Comput 58:211–222
Bluskov I, Heinrich K (2001) Super-simple designs with \(v\le 32\). J Stat Plan Inference 95:121–131
Bluskov I, Magliveras S (2001) On the number of mutually disjoint cyclic designs and large sets of designs. J Stat Plan Inference 95:133–142
Thanks go to Jan De Heer for reading this material and successfully testing the metaheuristic and for his valuable input in the process.
Electronic supplementary material
Below is the link to the electronic supplementary material.
About this article
Cite this article
Bluskov, I. Problem dependent optimization (PDO). J Comb Optim 31, 1335–1344 (2016). https://doi.org/10.1007/s10878-014-9826-x