The optimization of ship voyage management energy efficiency

Lowering fuel consumption of ships against volatile fuel prices and greenhouse gas emissions resulted from international shipping are the challenges that the industry faces today. CO2 emissions from maritime industry accounts for a significant part of total global greenhouse gas (GHG) emissions. According to the International Maritime Organisation (IMO), the contribution from ships was estimated to be 1016 million tonnes average for the period 2007-2012 which make up approximately 3.1% of global carbon emissions. With the tripling of world trade, if no action is taken, these emissions are forecasted to increase by 50% - 250% until 2050. OECD also stated a similar level of prediction in the increase in CO2 emissions from shipping industry. On the other hand, shipping companies encounters with high risks as a result of increased fuel prices to maintain their competitive power in the market. The fuel cost represents a large amount of the total operating cost of a shipping company which is estimated to be 50% or even more than 60%. Consequently, shipping companies focus on energy-efficient procedures and operations for decreasing energy consumption in order to lower their management costs and thus maintain their competitive position in the market. In this respect, the amount of energy-savings by ships' fuel should be increased and energy consumption should be reduced as much as possible for the above mentioned economic pressure and international legislation. Ship energy efficiency measures propose various alternatives to ship owners and operators to lower fuel consumption and carbon emissions. The potential for fuel savings in shipping by 25% to 75% is achievable through more efficient operations of existing ships and increased energy efficiency in the design of newbuilds. IMO's Marine Environment Protection Committee (MEPC) adopted the addition of new regulations related energy efficiency of ships to MARPOL (International Convention for the Prevention of Pollution from Ships) Annex VI, as a new chapter (Chapter 4). By the way of this, since 1st of January 2013, all new ships have to comply with an Energy Efficiency Design Index (EEDI) and all ships have to carry a Ship Energy Efficiency Management Plan (SEEMP).Moreover, Energy Efficiency Operational Indicator (EEOI) was recommended as a form of guidance to monitor the progress of the SEEMP. While EEDI suggests technology and design based measures at a minimum level with a long term impact for new ships, the aim of SEEMP is to enhance the energy efficiency through energy efficient ship operations using existing technologies on board a ship. Although there are many technology and design based approaches, limitations of these measures due to the long payback duration have led to discussion about the potential of implementing operational changes. For the above reasons, the fuel saving of ships has become paramount for ship energy efficiency. Decision support systems (DSS) is a computer-based approach that helps decision makers use data, models and other knowledge on the computer to solve semi structural and some non-structural problems, which cannot be measured or modelled. These problems requires human intervention, and therefore, solutions to semi-structured problems are often obtained by allowing a decision-maker to select and evaluate practical solutions the set of feasible alternatives. The goal of DSS is improving decision-making effectiveness and efficiency by Integrating of information sources and analysis tools. Combined effects of several factors are involved for evaluating ship voyage management energy efficiency measures. Providing a strategic approach to identify energy efficient solutions becomes more complicated for ship operators due to its complexity and difficulty. There is a need for decision support to provide quickly and directly solution for predicting fuel consumption at an operational level through implementing the most appropriate operational measures to increase energy efficiency against high oil prices and greenhouse emissions. The main aim of this study is twofold: (i) predict ship fuel consumption for various operational conditions through an inexact method, Artificial Neural Network ANN; (ii) develop a decision support system (DSS) employing ANN based fuel prediction model to be used on-board ships on a real time basis for voyage management energy efficiency optimization. Firstly, key operational energy efficiency measures within the scope of SEEMP have been examined in great detail. Fuzzy analytic hierarchy process (FAHP) which includes expert judgements is used to generate and calculate the quantitative importance of each operational measures. By this way, the importance of ship voyage management has been highlighted as a major ship operational energy efficiency measure, as well as, the voyage management measures- speed optimization, autopilot improvements, weather routing – safe and energy efficient route selection and trim optimization- have been prioritized subjectively. Within the scope of this study, the information and data of ship fuel consumption are acquired from mainly from noon reports and also supported by ships particulars information of the tanker ships. Noon reports are a major indicator of the amount of fuel that ships consume in various loading, speed and weather conditions that can be utilised for development of the ship voyage management energy efficiency. Noon reports derived from shipping company are examined statistically to explore the effects of operational factors related voyage management on fuel consumption. Pearson's correlation coefficient is used to test the correlation between fuel consumption and operational parameters, which are used as measures of ship energy efficiency. The results of this study will provide the most appropriate operational measures related voyage performance management to create ANN based decision support system. The parameters considered for fuel prediction have been determined through applying both subjective and objective approaches which are BAHP and statistical analyses, respectively. The results of these approaches reveal that reducing the speed of the ship is the most efficient method in terms of fuel economy and environmental impact. Although BAHP only assigns the weights of importance to the speed optimisation and trim optimization, correlation analyses also reveal a small but statistically significant correlation between fuel consumption and weather optimization. As a result, the fuel prediction model uses five input variables; revolutions per minute (RPM), mean draft, trim, (Beaufort number) BN and relative wind direction, in which output data is fuel consumption in metric ton per hour. The first main contribution is that this study presents the development of ships' fuel consumption prediction model based on artificial neural networks (ANN) model. ANN, one of the best efficient computational methods simulating the working principles of human brain, can achieve good solutions where the other traditional methods could not be effective. Several distinguishing features of ANN make them valuable for forecasting. First, it can learn from examples (past data) and capture subtle functional relationships among the data even if the underlying relationships are unknown or hard to describe. Second, ANN has generalization ability. After learning the data presented to them (a sample), ANN can often correctly infer the unseen part of a population even if the sample data contain noisy information. Third, ANN is nonlinear data-driven and it is capable for performing nonlinear modelling without an a priori knowledge about the relationships between input and outputs variables. The goal of the ANN is to predict ship fuel consumption under various operational conditions using operating data -'Noon Data' - which provides information on a ship's daily fuel consumption. The second main contribution of this study is to develop a decision support tool to help ship captain/operators making optimal decisions on a real time basis for energy efficient ship operations. By this way, ship captain / operator can manage the operational measures related voyage management depending on the speed optimization (RPM), draft optimization (trim and draft) and weather routing (BN and relative wind direction). The proposed method can be considered as a successful decision support tool for ship captain/operator in forecasting fuel consumption based on different daily operational conditions related voyage management energy efficiency. The proposed decision support system provides a strategic approach when ship operators have to make their decisions at an operational level considering both the economic and environmental aspects. The obtained results make it clear that the neural network can learn very accurately the relationships between the input variables and a ship's fuel consumption. As a further aspect, other decision support systems can be developed by using noon data derived from different types of ships such as bulk carriers, containers and etc. and larger number of ships with various characteristics. In addition, the scope of developed decision support systems can be extended by using other operational measures such as engine management and hull/propeller maintenance. Moreover, software engineers can structure the proposed decision support system as software package.