A Statistical Approach to Ship Fuel Consumption Monitoring

Abstract

The reduction of the environmental impact imposed by Kyoto Protocol and the growth of competitiveness imposed by the shipping market have urged shipping companies to pay increasing attention to ship energy efficiency improvement and CO2 emission reduction. According to the Ship Energy Efficiency Management Plan (SEEMP) recommended by the International Maritime Organization (IMO), the main scope of this article is in fact to overcome the deterministic limits of the monitoring systems installed in modern ships and support technical management in decision making based on large navigation databases. The proposed statistical approach is founded on a multiple linear regression and allows for both pointwise and interval predictions of the fuel consumption. At the end of each voyage, the model can be used to alert management for a possible change in ship performance in all those situations where the actual fuel consumption lies outside the prediction interval. Moreover, the model can also be utilized to quantify the contribution to the fuel consumption due to the hull and propeller fouling, which is particularly profitable for shipping companies and operators to claim for carbon credits after a specific improvement operation.

Data analytics for fuel consumption management in maritime transportation: Status and perspectives

“…However, as noon reports are completed manually, their accuracy cannot be guaranteed. Furthermore, only one record is made per day, which reduces the data quality and quantity (Bocchetti et al, 2013(Bocchetti et al, , 2015Erto et al, 2015;Chaal, 2018;Farag and Ölçer, 2020).…”

“…For modern ships equipped with onboard sensors, sensor data serve as another important data source (Petursson, 2009;Petersen et al, 2012aPetersen et al, , 2012bBocchetti et al, 2015;Jeon et al, 2018;Soner et al, 2018;Lepore et al, 2019;Capezza et al, 2019;Farag and Ölçer, 2020;Wang et al, 2016Wang et al, , 2020. To obtain various types of data, different types of sensors should be installed onboard.…”

“…Therefore, we expect that taking ship maintenance data into account in ship performance monitoring can make fuel consumption prediction more accurate and efficient. Nevertheless, as shown in Section 3, few ship fuel consumption prediction models take the time since last dry docking into consideration (Bocchetti et al, 2015;Erto et al, 2015;Rudzki and Tarelko, 2016;Adland et al, 2018Adland et al, , 2020, and other ship maintenance information, such as hull and propeller cleaning, main engine maintenance, and major overhaul information, is seldom considered.…”

Energy Consumption Optimization of Inland Sea Ships Based on Operation Data and Ensemble Learning

“…It is difcult to establish energy consumption prediction and optimization models with high accuracy to achieve the purpose of energy conservation and emission reduction of ships. In ship energy consumption prediction modeling, Bocchetti et al [11] used multiple linear regressions to model ship energy consumption and subsequently obtained a ftted equation for ship energy consumption per unit time versus speed. Soner et al [12] used the least absolute shrinkage and selection operator (LASSO) and the Ridge methods to model the ship energy consumption.…”

A functional data analysis approach for the monitoring of ship CO2 emissions

“…This typical high dimensionality issue cannot be overlooked by collapsing or averaging measurements when the aim of the analysis is to monitor or control the stability over time of quality characteristics apt to be modelled as a functional data. This approach has been used extensively in the literature (Bocchetti et al, 2015;Erto et al, 2015;Lepore et al, 2019;Capezza et al, 2019), however, there is a serious risk of discarding valuable information. For instance, Figure 1 shows two CO2 emission functions from the real-case study in the maritime field described in Section 3.…”