Uma abordagem heurística para um problema de rebalanceamento estático em sistemas de compartilhamento de bicicletas

• Ano de defesa: 2016
• Autor(a) principal: Albuquerque, Fabio Cruz Barbosa de
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal da Paraíba; Brasil; Informática; Programa de Pós-Graduação em Informática; UFPB
• Palavras-chave em Português: Metaheurísticas; Bike-sharing; Roteamento de veículos; Coleta e entrega com múltiplas visitas; Metaheuristics; Vehicle routing; Split pickup and delivery; Iterated local search; CIENCIAS EXATAS E DA TERRA::CIENCIA DA COMPUTACAO
• Link de acesso: https://repositorio.ufpb.br/jspui/handle/tede/9255

The Static Bike Rebalancing Problem (SBRP) is a recent problem motivated by the task of repositioning bikes among stations in a self-service bike-sharing systems. This problem can be seen as a variant of the one-commodity pickup and delivery vehicle routing problem, where multiple visits are allowed to be performed at each station, i.e., the demand of a station is allowed to be split. Moreover, a vehicle may temporarily drop its load at a station, leaving it in excess or, alternatively, collect more bikes (even all of them) from a station, thus leaving it in default. Both cases require further visits in order to meet the actual demands of such station. This work deals with a particular case of the SBRP, in which only a single vehicle is available and the objective is to nd a least-cost route that meets the demand of all stations and does not violate the minimum (zero) and maximum (vehicle capacity) load limits along the tour. Therefore, the number of bikes to be collected or delivered at each station should be appropriately determined in order to respect such constraints. This is a NP-Hard problem since it contains other NP-Hard problems as special cases, hence, using exact methods to solve it is intractable for larger instances. Several methods have been proposed by other authors, providing optimal values for small to medium sized instances, however, no work has consistently solved instances with more than 60 stations. The proposed algorithm to solve the problem is an iterated local search (ILS) based heuristic combined with a randomized variable neighborhood descent (RVND) as local search procedure. The algorithm was tested on 980 benchmark instances from the literature and the results obtained are quite competitive when compared to other existing methods. Moreover, the method was capable of nding most of the known optimal solutions and also of improving the results on a number of open instances.