An Evolutionary Algorithm For Gtsp Problem And Its Modelling

Zhifeng Hao,Han Huang,Xiaowei Yang
2007-01-01
Abstract:Generalized Travelling Salesman Problem, (GTSP) is one of the challenging combinatonal optimization, problems in a number of industrial applications. In general, GTSP is more complex than Travelling Salesman Problem (TSP). Recently, several methods such as heuristic algorithms and genetic algorithms have been developed for solving G TSP. Their performances, however, need to be improved. An evolutionary algorithm. (EA-GTSP) with hybrid encoding, special crossover operator, mutation operator and local optimization. strategy was proposed as the best method for GTSP. In this paper, EA-GTSP is modelled as a Markov Process and proved to converge to the global optimal solution. EA-GTSP has both global searching, which. is verified based on the analysis of Markov Process, and local accelerative optimal capacity shown in numerical results. All of the operators in EA-GTSP are studied theoretically and numerically to prove the effectiveness of EA-GTSP.
What problem does this paper attempt to address?