Maritime transportation has been continuously playing an undeniable role for the global trade and economy of many countries. Based on the fast growth of the maritime trade, the marine container terminal (MCT) operators should focus on improving the operations planning at the MCTs. Seaside operations have substantial impacts on the general throughput of the MCTs. The daily berth planning as a seaside operation is the point of focus herein. The daily berth planning is modeled as a berth scheduling problem (BSP) in this dissertation. The BSP (as a decision problem) aims to assign the arriving vessels to the available berthing positions and can be reduced to the unrelated machine scheduling problem, which has NP-hard complexity. The large-size instances of decision problems with NP-hard complexity cannot be solved using exact optimization algorithms, while metaheuristic algorithms can effectively solve large-size problem instances and return good-quality solutions. Evolutionary Algorithms (EAs) are among the most popular metaheuristic algorithms deployed to solve the real-size BSPs. There are some algorithmic parameters in EAs (e.g., crossover probability, mutation probability, population size, etc.), which should be assigned the appropriate values to have the best possible performance of the algorithm for a given BSP. The process of determination of algorithmic parameters values is called the parameter selection. Several methodologies have been introduced in the EA literature for parameter selection, which can be classified as follows: (1) parameter tuning; and (2) parameter control. In parameter tuning, the algorithmic parameter values remain constant throughout the algorithmic evolution, while the parameter control strategy updates the algorithmic parameters considering different approaches. In this dissertation, an EA with a self-adaptive parameter control strategy is proposed to solve the developed BSP. Based on a self-adaptive parameter control strategy, the crossover and mutation probabilities are encoded in the solutions and evolve with the EA. The problem is formulated as a mixed-integer linear programming model, minimizing the total weighted vessel turnaround time and the total weighted vessel late departures. Comprehensive numerical experiments are conducted to assess performance of the proposed self-adaptive EA against the alternative EAs, which rely on the different parameter selection strategies. Results demonstrate that all the considered solution algorithms show a promising performance in terms of the objective function values at termination. However, application of the self-adaptive parameter control strategy substantially enhances the objective function values at convergence without a significant impact on the computational time. Furthermore, an EA with an augmented self-adaptive parameter control strategy is presented in this dissertation as another solution algorithm for the BSPs. Based on an augmented self-adaptive parameter control strategy, not only the crossover and mutation probabilities are encoded in the solutions but they are also updated based on the feedback from the search. A mixed-integer linear programming mathematical model is developed for the BSP, aiming to minimize the total costs for serving vessels at the MCT. The designed algorithm is evaluated against nine alternative state-of-the-art metaheuristic algorithms, which have been widely utilized in the BSP literature. The results show that all the developed algorithms have a high level of stability and return high-quality solutions at termination. The computational experiments also prove the superiority of the designed augmented self-adaptive EA over the alternative algorithms considering different performance indicators. Another innovative solution methodology is developed in this dissertation, which relies on the island-based concept. Specifically, a universal island-based metaheuristic algorithm is designed for the BSP, where four different population-based metaheuristics are executed simultaneously in order to effectively search for solutions. A mixed-integer linear mathematical model is developed for the BSP, minimizing the total cost to serve the arriving vessels at the MCT. Comprehensive numerical experiments are conducted to evaluate performance of the island-based algorithm against seven commonly used metaheuristics in the BSP literature. The stability and the capability of the adopted algorithms in providing high-quality solutions at convergence are proven. The results demonstrate that the island-based algorithm outperforms other adopted algorithms considering different performance indicators. To summarize, this dissertation proposes three different solution methodologies for various BSP mathematical formulations. The algorithms have been evaluated based on extensive numerical experiments against the alternative algorithms, which have been widely used in the MCT and freight terminal operations literature. Findings confirm effectiveness of the proposed solution methodologies. Therefore, the developed solution methodologies can serve as promising decision support tools and assist MCT operators with the development of berth schedules. The latter will also assist with serving the growing demand for containerized trade and ensure that the vessel service will be completed in a timely manner. / A Dissertation submitted to the Department of Civil and Environmental Engineering in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / 2019 / October 29, 2019. / Berth Scheduling Problem, Marine Container Terminal, Metaheuristic, Optimization, Supply Chain / Includes bibliographical references. / Maxim A. Dulebenets, Professor Directing Dissertation; O. Arda Vanli, University Representative; Ren Moses, Committee Member; Eren Ozguven, Committee Member; Hui Wang, Committee Member.
| Identifer | oai:union.ndltd.org:fsu.edu/oai:fsu.digital.flvc.org:fsu_752365 |
| Contributors | Kavoosi, Masoud (author), Dulebenets, Maxim A. (professor directing dissertation), Vanli, Omer Arda (university representative), Moses, Ren (committee member), Ozguven, Eren Erman (committee member), Wang, Hui (committee member), Florida State University (degree granting institution), FAMU-FSU College of Engineering (degree granting college), Department of Civil and Environmental Engineering (degree granting departmentdgg) |
| Publisher | Florida State University |
| Source Sets | Florida State University |
| Language | English, English |
| Detected Language | English |
| Type | Text, text, doctoral thesis |
| Format | 1 online resource (216 pages), computer, application/pdf |
Page generated in 0.0025 seconds