One of the core challenges of the ship-owner companies nowadays is related to the reduction of fuel consumption for both environmental reasons due to the international regulations and for sustainability and economical reasons for the company itself. For the shipping industry, the fuel consumed per units (humans, cars, trucks) transported and of distance covered is one of the key measurements for the efficiency and profitability of the company.
The BigDataOcean platform aims at improving the business situation of ANEK by providing the means to investigate the impact of the environmental conditions and the operational decisions taken on the vessel’s fuel consumption. ANEK will utilise the BigDataOcean platform in order to investigate the possible correlation of the proprietary historical data containing information from the company’s passenger vessels with open historical meteorological and environmental data, as well as near-real-time meteorological and environmental data in regard to the fuel consumption and the possible impact on it.
Measures for the improvement of ship energy efficiency are normally divided into design and operational measures. The design measures include, but may not be limited to, a series of techniques and measures, such as the Air Lubrication System, energy-saving devices named Hi-FIN fuel and noise reduction systems called Onboard DV Grid systems and more, that contribute to the reduction of costs and offer a larger saving potential but require a significant capital investment from the ship owner’s. On the other hand, the operational measures that can be investigated and applied can produce significant reduction in fuel consumption with much more limited capital investment and contribute in the objectives of the ship-owner company in terms of cost minimisation, profitability and environmental impact reduction. Under the umbrella of operational measures, different measures for energy efficiency on board are considered, that do not require the installation of new equipment, and include among others, the following: a) improvement in voyage execution, b) reduction of auxiliary power consumption, c) initial fuel and diesel loads, initial water and lubricants loads, d) weather routing, e) optimised hull and propeller polishing schedule, f) slow steaming, and g) trim optimisation. As it is obvious, the number of variables that they influence the ship’s energy efficiency is quite large and the assessment of the ship’s performance in relation to a standard baseline is rather difficult. Although the energy demand for propulsion represents the largest share of the total energy demand, this is tightly related and influenced by a large number of variables such as the current weather conditions and various operational choices by the vessel’s captain such as the trim of the vessel and the vessel’s speed. Thus, an appropriate optimisation requires an in-depth understanding of the influence of the speed of the vessel on its fuel consumption in different environmental and operational conditions.
Towards this end, the scenario aims at: (a) correlating the significant amount of historical data related with the draft of the vessel (directly influenced by the initial loads of fuel, diesel, water and lubricants as well as by the cargo (humans, cars and trucks) loaded) as provided by ANEK’s systems, with accurate historical weather data that are provided by open sources (Copernicus and Open Weather APIs) in order to identify the correlations between those parameters, (b) exploiting the more accurate near-real-time short term weather predictions as provided by open sources (Copernicus and Open Weather APIs) to estimate the impact that the operational decisions taken (vessel speed in knots) will have on the fuel consumption of the vessel.
To achieve this, data analytics techniques will be applied towards the analysis of:
• Data from EPOS (Operation System) of ANEK that incorporates all the data from the passenger vessels of the company with information, among others, related to the course of the vessels, as well as fuel and trip along with the corresponding metadata. In detail, the vessel data and their metadata include the identifier of the ship, the latitude and longitude of its position, the current speed and direction, as well as the timestamp of the record entry. Additionally for the vessel trip data, metadata are included concerning the identifier of the trip, the identifier of the ship and the starting and final terminal of the trip. Finally, metadata are included for the current levels of fuels, water, diesel, and lubricants, as well as the fuel, water, diesel and lubricants consumed during the trip, the number of passengers, cars, trucks during the trip.
- Historical weather data retrieved from open sources (Copernicus and Open Weather APIs) containing information with regards to horizontal wind speed, gust wind speed and significant wave height.
- Near-real-time short term weather forecasts retrieved from open sources (Copernicus and Open Weather APIs) containing information with regards to weather, sea-state, horizontal wind speed, gust wind speed and significant wave height.
The execution of the described scenario is composed by the following two test cases:
- P1SC2_1 – Fuel consumption reduction investigation preparation: In this first test case, the desired fuel consumption reduction strategies will be tested, in which the correlation of the historical data related to the draft of the vessel and the historical weather data obtained from open sources is exploited.
- P1SC2_2- Fuel consumption reduction investigation real-time execution: In this test case, the impact that the operational decisions taken will have on the fuel consumption of the vessel will be estimated, utilising the more accurate near-real-time short term weather predictions.
In accordance with the demonstrators implementation plan, the first phase of both test cases (P1SC2_1 and P1SC2_2) was implemented and integrated with the BigDataOcean platform. Both test cases were executed following the steps described above and an assessment was performed under real circumstances. The scope of the first evaluation was to provide the initial feedback for the first phase of both test cases of this scenario and through this evaluation to extract useful insights that will be translated into improvements and adjustments on both the development of the service, but also the interactions with the BigDataOcean platform in general.
The initial collected results are indicating the improvement of the envisaged business situation, even though this is the initial version of the service. From the produced results, valuable knowledge was extracted and the basis was set for the second and final implementation phase were more sophisticated fuel consumption reduction strategies will be tested and the impact of the operational decisions taken in correlation with the near-real-time short term weather predictions will be further evaluated.
The valuable feedback and evaluation collected indicate that a series of improvements and adjustments should introduced in the second phase of the implementation of both test cases. The aim in this phase will be to increase the perceived usefulness and the perceived ease of use of the service in order to fulfil the requirements of the stakeholders, as well as to increase the added value offered by the service. As such, the implementation activities of the second phase will be driven from the refinements that are required on the existing functionalities, as well as from the new functionalities that will be introduced.