Abstract
A unique integer linear programming formulation is proposed to define a Hamiltonian tour in which the cost between a node pair is determined by the number of nodes between each pair on the tour. In addition, a node separation requirement may be conditional on other node separations. Properties of this new traveling salesman problem formulation are discussed, along with an a priori solution to the probabilistic traveling salesman problem with equal node coverage probability as a linear programming alternative to the probabilistic programming method in the literature. A linear programming-based heuristic is proposed and numerically tested.
Get full access to this article
View all access options for this article.
