Informatica

In this paper, an efficient hybrid genetic algorithm (HGA) and its variants for the well-known combinatorial optimization problem, the quadratic assignment problem (QAP) are discussed. In particular, we tested our algorithms on a special type of QAPs, the structured quadratic assignment problems. The results from the computational experiments on this class of problems demonstrate that HGAs allow to achieve near-optimal and (pseudo-)optimal solutions at very reasonable computation times. The obtained results also confirm that the hybrid genetic algorithms are among the most suitable heuristic approaches for this type of QAPs.

Recently, genetic algorithms (GAs) and their hybrids have achieved great success in solving difficult combinatorial optimization problems. In this paper, the issues related to the performance of the genetic search in the context of the grey pattern problem (GPP) are discussed. The main attention is paid to the investigation of the solution recombination, i.e., crossover operators which play an important role by developing robust genetic algorithms. We implemented seven crossover operators within the hybrid genetic algorithm (HGA) framework, and carried out the computational experiments in order to test the influence of the recombination operators to the genetic search process. We examined the one point crossover, the uniform like crossover, the cycle crossover, the swap path crossover, and others. A so-called multiple parent crossover based on a special type of recombination of several solutions was tried, too. The results obtained from the experiments on the GPP test instances demonstrate promising efficiency of the swap path and multiple parent crossovers.

In this paper we present an algorithm for generating quadratic assignment problem (QAP) instances with known provably optimal solution. The flow matrix of such instances is constructed from the matrices corresponding to special graphs whose size may reach the dimension of the problem. In this respect, the algorithm generalizes some existing algorithms based on the iterative selection of triangles only. The set of instances which can be produced by the algorithm is NP-hard. Using multi-start descent heuristic for the QAP, we compare experimentally such test cases against those created by several existing generators and against Nugent-type problems from the QAPLIB as well.

Many heuristics, such as simulated annealing, genetic algorithms, greedy randomized adaptive search procedures are stochastic. In this paper, we propose a deterministic heuristic algorithm, which is applied to the quadratic assignment problem. We refer this algorithm to as intensive search algorithm (or briefly intensive search). We tested our algorithm on the various instances from the library of the QAP instances – QAPLIB. The results obtained from the experiments show that the proposed algorithm appears superior, in many cases, to the well-known algorithm – simulated annealing.

In this paper we define a class of edge-weighted graphs having nonnegatively valued bisections. We show experimentally that complete such graphs with more than three vertices and also some special graphs with only positive edges can be applied to improve the existing lower bounds for a version of the quadratic assignment problem, namely with a matrix composed of rectilinear distances between points in the Euclidean space.