Sustainable Vehicle Routing Models with Mixed Fleet Vehicles
| dc.contributor.author | Islam, Md Anisul | |
| dc.contributor.examiningcommittee | Dr. Qingjin Peng (Mechanical Engineering), Dr. Srimantoorao S. Appadoo (Supply Chain Management) | en_US |
| dc.contributor.supervisor | Dr. Tarek ElMekkawy (Mechanical Engineering), Dr. Yuvraj Gajpal (Supply Chain Management) | en_US |
| dc.date.accessioned | 2021-01-12T20:06:36Z | |
| dc.date.available | 2021-01-12T20:06:36Z | |
| dc.date.copyright | 2021-01-04 | |
| dc.date.issued | 2020-12-22 | en_US |
| dc.date.submitted | 2021-01-05T04:07:48Z | en_US |
| dc.degree.discipline | Mechanical Engineering | en_US |
| dc.degree.level | Doctor of Philosophy (Ph.D.) | en_US |
| dc.description.abstract | In recent years, decarbonization of transportation has aroused great attention in the world. The transportation sector generates approximately one-fourth of global CO2 emission yearly and the amount is predicted to be double in 2050. To deal with this issue, sustainable development is forcing transportation sectors to minimize carbon emission produced by the cargo fleets. This thesis introduces vehicle routing models with a mixed fleet to deal with the sustainable development effort of the transportation industry. The mixed fleet in VRP models consists of heterogeneous hydrogen vehicles and conventional vehicles for the distribution system. The hydrogen vehicle has been introduced as an alternative fuel vehicle (AFV) in vehicle routing models for the first time in this work. The fuel consumption of the vehicles is realistically considered as a function of traveled distance, speed, and on-board cargo load. The problems include constraints of vehicle capacity, backhaul, time windows, CO2 emission cap, and maximum tour length for the routes. The thesis studies three VRP models. A new hybrid metaheuristic, combining the particle swarm optimization (PSO) and problem specific variable neighborhood search (VNS), is proposed to solve each of the investigating problems. Firstly, it considers a clustered vehicle routing problem (CluVRP), where customers are grouped into different clusters. A vehicle visiting a cluster cannot leave the cluster until all customers in the same cluster have been served. Each cluster and customer has to be served only once. The proposed hybrid PSO algorithm is tested on benchmark instances obtained from the CluVRP literature. The thesis then considers the mixed fleet green VRP with backhaul and time windows (MFGVRPTW). Extensive computational experiments have been performed on newly generated instances and benchmark instances with various sizes obtained from the literature of VRPB, VRPTW, and VRPBTW. Finally, a comprehensive VRP model called green clustered VRP with backhaul and time windows (GCluVRPBTW) is developed for the first time in this thesis. The performance of the proposed hybrid algorithm is evaluated by testing on newly generated GCluVRPBTW, CluVRPB, CluVRPTW, and CluVRPBTW. The proposed algorithm proved to be superior to the state-of-the-art algorithms on the CluVRP, VRPB, VRPTW, and VRPBTW. | en_US |
| dc.description.note | February 2021 | en_US |
| dc.identifier.citation | APA | en_US |
| dc.identifier.uri | http://hdl.handle.net/1993/35213 | |
| dc.language.iso | eng | en_US |
| dc.rights | open access | en_US |
| dc.subject | Mixed fleet Green VRPBTW, Green clustered VRPBTW, Clustered VRP, Hybrid PSO, Sustainability, CO2 emission | en_US |
| dc.title | Sustainable Vehicle Routing Models with Mixed Fleet Vehicles | en_US |
| dc.type | doctoral thesis | en_US |