Analysis of the dynamic traveling salesman problem with weight changes

In the dynamic traveling salesman problem (TSP), the vertices and/or weights of the graph that represents the TSP are changed during the optimization process. The dynamic TSP with weight changes is investigated. The effects of the changes on the fitness landscapes of the problem are analyzed. Questions regarding the dynamic TSP, like "how many solutions are affected by a change?" and "how does the severity of the problem influence the optima?", are discussed. Simulations of the dynamic TSP with weight changes are presented and analyzed. In the simulations, it is possible to observe that the new best solutions after a change are generally not far from the old best solutions. Also, it is possible to observe that there is a threshold for the maximum value of severity of the problem. Beyond this threshold, the changes have the same impact on the fitness landscape.