The pollution traveling salesman problem (PTSP) and the energy minimization traveling salesman problem (EMTSP) generalize the well-known asymmetric traveling salesman problem by including environmental issues and the goal of reducing carbon emissions. Both problems call for determining a Hamiltonian tour that, in the PTSP, minimizes a function of fuel consumption and driver cost (where the fuel consumption depends on the distance traveled, the vehicle speed, and the vehicle load), while, in the EMTSP, minimizes a function depending on the vehicle load and the traveled distances. For both PTSP and EMTSP, we propose a matheuristic algorithm that uses the solution of the linear programming relaxation of a mixed integer linear programming model for the considered problem to determine good initial feasible solutions, applies a multioperator genetic algorithm to improve these solutions, and refines the best solution found through an iterated local search procedure. In order to evaluate the performance of the proposed matheuristics, we compare them with exact and heuristic algorithms from the literature on benchmark instances of both problems.
A matheuristic algorithm for the pollution and energy minimization traveling salesman problems
Cacchiani V.Primo
;Contreras-Bolton C.;Toth P.
2023
Abstract
The pollution traveling salesman problem (PTSP) and the energy minimization traveling salesman problem (EMTSP) generalize the well-known asymmetric traveling salesman problem by including environmental issues and the goal of reducing carbon emissions. Both problems call for determining a Hamiltonian tour that, in the PTSP, minimizes a function of fuel consumption and driver cost (where the fuel consumption depends on the distance traveled, the vehicle speed, and the vehicle load), while, in the EMTSP, minimizes a function depending on the vehicle load and the traveled distances. For both PTSP and EMTSP, we propose a matheuristic algorithm that uses the solution of the linear programming relaxation of a mixed integer linear programming model for the considered problem to determine good initial feasible solutions, applies a multioperator genetic algorithm to improve these solutions, and refines the best solution found through an iterated local search procedure. In order to evaluate the performance of the proposed matheuristics, we compare them with exact and heuristic algorithms from the literature on benchmark instances of both problems.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
