Network design with route planning for battery electric high-speed passenger vessel services
Håkon Furnes Havre, Ulrik Lien, Mattias Myklebust Ness, Kjetil Fagerholt, Kenneth Løvold Rødseth
Abstract
This paper studies the Zero Emission passenger Vessel Service Network Design Problem (ZEVSNDP) in order to investigate how technical and economic challenges related to diffusion of battery electric vessels can be alleviated by appropriate planning of services. The ZEVSNDP considers decisions that are strategic (i.e., vessel fleet and charging locations), tactical (i.e., routes, whether to omit servicing ports, fleet deployment, and operating frequencies), as well as operational (i.e., passenger flow, sailing speeds, and scheduling decisions). A novel Mixed Integer Programming (MIP) model considering operator and passenger costs is proposed for the ZEVSNDP. Since the MIP model cannot be solved to optimality by a commercial solver except for tiny instances, we implement a heuristic Decomposition Based (DB) solution method. The DB solution method is applied to a real complex passenger vessel service in Florø, Norway, as well as two other test instances focusing on short-range transport and dense markets, respectively. Except for the short-range test instance, abatement costs (i.e., the costs of removing CO2 emissions by introducing battery electric vessels) are found to be significant. This is attributed to limited reach and time used for charging of battery electric vessels. Routes should consequently accommodate range limitations: omitting ports from the current route can be a cost-effective strategy when the cost of alternative transport for the passengers is moderate.