Competitive memetic algorithms for arc routing problems -P. Lacomme et al.
RR LOSI-2001-01 -Page 1
Competitive Memetic Algorithms for Arc Routing ProblemsPhilippe LACOMME
AbstractThe Capacitated Arc Routing Problem or CARP arises in applications like waste collection or winter gritting. Metaheuristics are tools of choice for solving large instances of this NP-hard problem. The paper presents basic components that can be combined into powerful memetic algorithms (MAs) for solving an extended version of the CARP (ECARP). The best resulting MA outperforms all known heuristics on three sets of benchmark files containing in total 81 instances with up to 140 nodes and 190 edges. In particular, one open instance is broken by reaching a tight lower bound designed by Belenguer and Benavent, 26 best-known solutions are improved, and all other best-known solutions are retrieved.Keywords: Capacitated Arc Routing Problem, CARP, metaheuristic, memetic algorithm.Competitive memetic algorithms for arc routing problems -P. Lacomme et al.
We consider here a logistic platform or more generally a node of a supply chain. After previous research works at the planning level whose aim was to smooth the workload by modifying slightly arrival and departure dates, we are now interested by the scheduling level. Our particular industrial framework led us to original hypotheses: given component quantities are delivered by trucks at some fixed times; a first optimized tour of the customers is planned at a known fixed date and a second optimized tour will be executed at a flexible date corresponding to the end of the schedule with the remaining customer orders. We reduce the activity inside the platform to the most important operation. This operation is performed by a single non renewable resource. Nevertheless most of the presented results could be easily extended to identical parallel machines. The considered scheduling problem is NP-Hard. With the goal of solving it by a branch and bound approach, we propose here a series of upper bounds (rapid approximation methods) and a series of lower bounds (obtained by various relaxations). Experimentations permit us to compare quality and computational times of the lower bounds and give us a first idea of the quality of the rapid approximation approaches.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.