Search

The p-median problem is a well-known NP-hard problem. Many heuristics have been proposed in the literature for this problem. In this paper, we exploit a GPGPU parallel computing platform to present a new genetic algorithm implemented in Cuda and based on a Pseudo Boolean formulation of the p-median problem. We have tested the effectiveness of our algorithm using a Tesla K40 (2880 Cuda cores) on 290 different benchmark instances obtained from OR-Library, discrete location problems benchmark library, and benchmarks introduced in recent publications. The algorithm succeeded in finding optimal solutions for all instances except for two OR-library instances, namely pmed30 and pmed40, where better than 99.9\% approximations were obtained.

There is a sizable literature on investigating the minimum and maximum numbers of cycles in a class of graphs. However, the answer is known only for special classes. This paper presents a result on the smallest number of cycles in hamiltonian 3-connected cubic graphs. Further, it describes a proof technique that could improve an upper bound of the largest number of cycles in a hamiltonian graph., Comment: Accepted for publication in Mathematica Slovaca

Lee codes have been intensively studied for more than 40 years. Interest in these codes has been triggered by the Golomb-Welch conjecture on the existence of the perfect error-correcting Lee codes. In this paper we deal with the existence and enumeration of diameter perfect Lee codes. As main results we determine all $q$ for which there exists a linear diameter-4 perfect Lee code of word length $n$ over $Z_{q},$ and prove that for each $n\geq 3$ there are uncountable many diameter-4 perfect Lee codes of word length $n$ over $Z.$ This is in a strict contrast with perfect error-correcting Lee codes of word length $n$ over $Z\,$\ as there is a unique such code for $n=3,$ and its is conjectured that this is always the case when $2n+1$ is a prime. We produce diameter perfect Lee codes by an algebraic construction that is based on a group homomorphism. This will allow us to design an efficient algorithm for their decoding. We hope that this construction will turn out to be useful far beyond the scope of this paper.