Improving search by incorporating evolution principles in parallel Tabu Search

摘要: Combinatorial optimization problems require computing efforts which grow at least exponentially with the problem dimension. Therefore, use of remarkable power massively parallel systems constitutes an opportunity to be considered for solving significant applications in reasonable times. In this paper, starting from Tabu Search, a general methodology, version, oriented distributed memory multiprocessors and including evolution principles, has been introduced discussed. The experiments have performed on classical Traveling Salesman Problems Quadratic Assignment taken literature. results obtained show that incorporation principles is very fruitful search strategy terms both convergence speed solution precision. >