Premium
On the design of hybrid bio‐inspired meta‐heuristics for complex multiattribute vehicle routing problems
Author(s) -
Nogareda AnaMaria,
Del Ser Javier,
Osaba Eneko,
Camacho David
Publication year - 2020
Publication title -
expert systems
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.365
H-Index - 38
eISSN - 1468-0394
pISSN - 0266-4720
DOI - 10.1111/exsy.12528
Subject(s) - ant colony optimization algorithms , computer science , benchmark (surveying) , heuristics , genetic algorithm , vehicle routing problem , mathematical optimization , memetic algorithm , metaheuristic , local search (optimization) , routing (electronic design automation) , heuristic , computation , algorithm , artificial intelligence , machine learning , mathematics , computer network , geodesy , geography , operating system
This paper addresses a multiattribute vehicle routing problem, the rich vehicle routing problem, with time constraints, heterogeneous fleet, multiple depots, multiple routes, and incompatibilities of goods. Four different approaches are presented and applied to 15 real datasets. They are based on two meta‐heuristics, ant colony optimization (ACO) and genetic algorithm (GA), that are applied in their standard formulation and combined as hybrid meta‐heuristics to solve the problem. As such ACO‐GA is a hybrid meta‐heuristic using ACO as main approach and GA as local search. GA‐ACO is a memetic algorithm using GA as main approach and ACO as local search. The results regarding quality and computation time are compared with two commercial tools currently used to solve the problem. Considering the number of customers served, one of the tools and the ACO‐GA approach outperforms the others. Considering the cost, ACO, GA, and GA‐ACO provide better results. Regarding computation time, GA and GA‐ACO have been found the most competitive among the benchmark.