Waiting strategies for the Dynamic and Stochastic Traveling Salesman Problem