Premium
An evolutionary algorithm for the multi‐objective pick‐up and delivery pollution‐routing problem
Author(s) -
Bravo Mauricio,
Rojas Lorena Pradenas,
Parada Victor
Publication year - 2019
Publication title -
international transactions in operational research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.032
H-Index - 52
eISSN - 1475-3995
pISSN - 0969-6016
DOI - 10.1111/itor.12376
Subject(s) - vehicle routing problem , mathematical optimization , computer science , evolutionary algorithm , constraint (computer aided design) , pareto principle , benchmark (surveying) , greenhouse gas , minification , routing (electronic design automation) , operations research , mathematics , ecology , computer network , geometry , geodesy , biology , geography
The design of sustainable logistics solutions poses new challenges for the study of vehicle‐routing problems. The design of efficient systems for transporting products via a heterogeneous fleet of vehicles must consider the minimization of cost, emissions of greenhouse gases, and the ability to serve every customer within an available time slot. This phenomenon gives rise to a multi‐objective problem that considers the emission of greenhouse gases, the total traveling time, and the number of customers served. The proposed model is approached with an ε ‐constraint technique that allows small instances to be solved and an evolutionary algorithm is proposed to deal with complex instances. Results for small instances show that all the points that approach the Pareto frontier found by the evolutionary algorithm are nondominated by any solution found by the multi‐objective model. For complex instances, nondominated solutions that serve most of the requests are found with low computational requirements.