Development of a statistical model for forecasting fuel consumption with application to ship route assessment

Αποστόλου, Δημήτρης; Apostolou, Dimitris

dc.contributor.author	Αποστόλου, Δημήτρης	el
dc.contributor.author	Apostolou, Dimitris	en
dc.date.accessioned	2022-01-20T10:58:43Z
dc.date.available	2022-01-20T10:58:43Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/54394
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.22092
dc.rights	Default License
dc.subject	Παλινδρόμηση	el
dc.subject	Μοντέλο	el
dc.subject	Κατανάλωση καυσίμου	el
dc.subject	Αξιολόγηση διαδρομής	el
dc.subject	Πρόβλεψη	el
dc.subject	Weather routing	en
dc.subject	Regression	en
dc.subject	Fuel consumption	en
dc.title	Development of a statistical model for forecasting fuel consumption with application to ship route assessment	en
heal.type	bachelorThesis
heal.classification	Ship performance	en
heal.language	en
heal.access	campus
heal.recordProvider	ntua	el
heal.publicationDate	2021-07-19
heal.abstract	In the present diploma thesis a data-driven model is developed, which predicts fuel oil consumption for a crude oil tanker based on specific voyage’s parameters and gives attention to route-related factors. The required dataset for the construction of the model is based on measurements obtained by an onboard monitoring system. In order to acquire a reliable dataset, a pre-processing procedure takes place, which includes the removal of outlying data-points through imposition of thresholds and statistical filtering. After analyzing the relation between the dependent and the independent variables, the model is developed using multiple linear regression and the accuracy it offers is satisfactory. The conducted case study, focusing on weather routing, reveals that for the current model, fuel consumption is hardly affected by the encountered wind and that, under certain conditions, a route longer in length (compared to the shortest possible) but with more favorable current might be preferable, leading to noteworthy fuel savings. It is recommended, however, that a non-linear model shall be developed for more accurate depiction of the problem’s physics.	en
heal.advisorName	Θεμελής, Νικόλαος	el
heal.committeeMemberName	Σπύρου, Κωνσταντίνος	el
heal.committeeMemberName	Ζαραφωνίτης, Γεώργιος	el
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ναυπηγών Μηχανολόγων Μηχανικών. Τομέας Μελέτης Πλοίου και Θαλάσσιων Μεταφορών	el
heal.academicPublisherID	ntua
heal.numberOfPages	87 σ.	el
heal.fullTextAvailability	false