Tabu Search with Multiple Decision Levels for Solving Heterogeneous Fleet Pollution Routing Problem

Abstract

Organizations, in order to gain a competitive advantage, must improve their logistics performance along with the planning and distribution of their goods. Thus, they face significant challenges in managing their orders to be delivered on time. However, transportation is responsible for of the emissions of the total polluting gases in the atmosphere. Therefore, there is a growing interest to investigate methods to optimize logistics and to consider environmental aspects. However, the literature only considers realistic system characteristics such as: different vehicles and speeds, time windows and route inclination. For this reason, the focus is on the solution of an extension with a heterogeneous fleet and discrete speeds of the Vehicle Routing Pollution Problem (PRP), whose objective is the reduction of greenhouse gases (GHG). Based on the MEET model, the main polluting gases with the greatest impact on health are measured: carbon dioxide, nitrogen dioxide and carbon monoxide (CO). For its solution, a Tabu Search metaheuristic is proposed with different decision levels: node sequence, assigned speeds and vehicles used, from different neighborhood structures. Finally, the balance between exploration and exploitation is achieved by incorporating favorable attributes to the created solutions. The proposed metaheuristic achieves efficient results both in total logistic cost and in emissions released to the environment.