Classification societies’ regulations on shaft alignment

Μπερερή, Χριστίνα; Bereri, Christina

dc.contributor.author	Μπερερή, Χριστίνα	el
dc.contributor.author	Bereri, Christina	en
dc.date.accessioned	2024-12-10T08:57:01Z
dc.date.available	2024-12-10T08:57:01Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/60516
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.28212
dc.rights	Default License
dc.subject	Ευθυγράμμιση	el
dc.subject	Αξονικά	el
dc.subject	Νηογνώμονας	el
dc.subject	Κανονισμοί	el
dc.subject	Alignment	en
dc.title	Classification societies’ regulations on shaft alignment	en
dc.title	Κανονισμοί νηογνωμόνων για την ευθυγράμμιση αξονικών συστημάτων πλοίου	el
heal.type	bachelorThesis	el
heal.classification	Ναυτική μηχανολογία	el
heal.language	en	el
heal.access	free	el
heal.recordProvider	ntua	el
heal.publicationDate	2024-07-08
heal.abstract	Proper Shaft Alignment ensures the uninterrupted and optimal operation of a vessel. Erroneous shaft alignment has serious consequences, including malfunctions and significant damage to a ship's shafting system. The implications could range from vessel service interruptions and property loss to environmental disasters. As a result, it is critical that all parties responsible for the design, construction, operation, and monitoring of vessels take all required procedures and precautions to eliminate the possibility of the aforementioned undesirable situations occurring. The Shaft Alignment procedure of a vessel is a complex and multidimensional matter, as the responsible engineers are asked to regulate and fine-tune numerous parameters of the Shafting System, which are dependent to one another, while the entire Shafting System is significantly sensitive to several characteristics of the vessel's structure and operating profile. Furthermore, the Shafting System necessitates careful handling while taking measurements or installing components and mechanisms, as clearances are on the order of a few millimeters. Classification Societies are responsible for evaluating and verifying compliance to high standards that ensure the vessel’s safety and seaworthiness during its entire lifespan, from design to demolition. Classification Societies ensure that all mechanisms, construction materials, structure components, design characteristics, and safety protocols guarantee the vessel's safe operation throughout its lifecycle, thereby protecting property, the environment, and life at sea. The creation and publication of Shaft Alignment Rules and Regulations falls within the purview of a Classification Society's responsibilities. However, each IACS Class has its own documentation strategies, terminology, and scientific approach to relatively similar Regulations. This phenomenon also applies to Shaft Alignment Regulations, making the process of collecting and comparing Regulations from various Classification Societies for professional, educational or comprehensive purposes challenging. One of the main ideas that stimulated this thesis is that during their studies, Naval Architecture and Marine Engineering students are frequently asked to gather, understand, apply, and compare the Rules and Regulations of various Classification Societies. This thesisis divided into two major parts: The first part explains and discusses the fundamental principles and characteristics of a shafting system. The rules and regulations of several IACS member Classes (ABS, DNV, LR,BV, and ClassNK) are then documented, explained, and compared. The Rules and Regulations used for this thesis were the most recent updates that each Classification Society publicly provided. The variations and similarities between the regulations released by the aforementioned Classes in terms of technical approach, terminology used, limits set up, suggested tools, recommended practices, and methodologies are then highlighted and analyzed. This work may serve as a guide for educational purposes in the future by providing extensive explanation of the theoretical and operational principles, the main components, and the respective details of rules and regulations regarding shaft alignment in marine vessels, providing students with a comprehensive view of the matter. Furthermore, Artificial intelligence (AI) is continually evolving, and it is increasingly making its way into our daily lives and important corporate sectors. In terms of document classification and management, AI is commonly used to handle, search, and analyze enormous amounts of data for educational or business objectives. Additionally, with the increased digitization of documents, the use of AI technology is becoming increasingly necessary. Large Language Models (LLMs) are thought to be a promising option because they can interpret and handle text, as well as generate human-like text and conduct language-related tasks. The Second Part of this thesis is focused on integrating LLMs into a comprehensive document management system capable of understanding and responding correctly to specific technical queries regarding the marine engineer’s profession. In the context of this thesis, experimentation was conducted with CHAT-GPT 3.5 to acquire a better understanding of how this advanced LLM, replied to a series of technical questions about “Shaft Alignment” and the applicable Regulations provided by Classification Societies. Studying and categorizing the answers provided by CHAT-GPT 3.5 focusing mainly on the answers that the AI model answered incorrectly or inadequately, a pool of questions of interest was created. These questions were then addressed to other opensource LLMs (like Llama 3, Nous Hermes 2 Mistral DPO and GPT4ALL Falcon), which had local document access to the rules, or their comparative summary presented in Part A. All of the questions, which were eventually incorporated in the final set utilized within this thesis’, represent a specific “Quality” Category regarding the performance of each LLM that was tested. Finally, the best performing open source LLM can be further enhanced via Fine-Tuning and be utilized for Academic, Educational or professional purposes, in the context of providing an AI model capable of answering technical questions specified in Shaft Alignment and the respective Classification Societies’ Regulations. Finally, the outcomes of this process may indicate that LLMs that are provided access in appropriate (Local) Documents or fine-tuned with targeted Datasets can be used for more advanced applications.	en
heal.advisorName	Παπαδόπουλος, Χρήστος	el
heal.committeeMemberName	Ανυφαντής, Κωνσταντίνος	el
heal.committeeMemberName	Παπαλάμπρου, Γεώργιος	el
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ναυπηγών Μηχανολόγων Μηχανικών. Τομέας Ναυτικής Μηχανολογίας	el
heal.academicPublisherID	ntua	el
heal.numberOfPages	196 σ.	el
heal.fullTextAvailability	false