|dc.description.abstract||Up until the financial crisis in 2008, container liner operators traditionally utilized economies of scale by building larger and faster vessels, with the main focus of maximizing profit. Today, overcapacity and higher fuel prices force liner operators to cut costs and develop new strategies to survive. This thesis investigates the possibility of deploying electric and autonomous container feeders in short sea shipping networks. An approach utilizing mathematical programming and methodologies from systems engineering is developed to investigate the performance of various fleet compositions in possible future scenarios. Routing optimization determines the optimal fleet operation, and is combined with tradespace exploration providing high-level observations into fleet performance in different contexts.
A computational study is conducted with the goal of identifying scenarios in which fleets comprised of electric or autonomous vessels outperform conventional container feeders in terms of cost and time efficiency. First, a set of variables related to exogenous factors that influence the cost of operating and building electric, autonomous and diesel vessels are determined. These variables define the context of the future scenarios, or epochs, investigated in this thesis. The ship design process generates 23 differently sized vessels each type, and is based a parametric assessment of existing container feeders. The design of ships with electric propulsion systems is also dependent on the size of the battery packs they require. In total, 140 fleets are generated, both homogeneous and heterogeneous with respect to ship size, comprised of either diesel, electric or autonomous vessels.
Two short sea transportation tasks are investigated, both with planning horizons of one week. The first is a hub-spoke scenario, and the second consists of randomly distributed cargoes between all ports. A cargo routing and scheduling formulation with time windows and split loads is presented, and determines each fleet's optimal routing mode by minimizing sailing cost. Finally, tradespace plots are developed for each epoch, illustrating the fleets design characteristics, costs and time efficiency.
The study found that a 87\% decrease in electricity prices in combination with 33\% higher MDO prices is required for electric and autonomous fleets to significantly outperform diesel fleets in terms of running costs. The cost of building large ships with electric propulsion systems remained significantly higher than the cost of build diesel vessels throughout the epochs investigated, due to costly battery packs. It was found that when subjected to the second transportation task, fleets comprised of a high number of smaller ships were more time efficient than fleets comprised of fewer, larger vessels. Conclusively, the results indicate that given advancements within battery technology, electric and autonomous ships may be able to outperform diesel ships in scenarios where small ships are required to sail short distances.||en