Data-driven bunker planner system

Abstract

The present invention describes a data-driven bunker planner system (11) for optimizing bunkering costs for a marine vessel (1). The bunker planner system (11) runs on a computing system (10) located on the vessel (1) and includes a data acquisition module (111), a fuel price forecast module (112) and a data optimization module (113). The fuel price forecast module (112) forecasts future fuel prices and / or fuel indices in the long or short term. The data optimization module (113) determines the optimal time to execute a bunkering term contract or combination of contracts based on the forecasted fuel prices and / or fuel indices. The forecasted fuel prices and / or fuel indices, the optimized fuel term contract, and the bunkering plan are output through a notification generation module (114) and a human-machine interface system (18). The bunker planner system output has been validated by using real data trained on various customized machine learning and optimization models.

Description

[Technical field]

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS The present invention claims priority to Singapore Patent Application No. 10202203430W, filed April 4, 2022, the disclosure of which is incorporated herein by reference in its entirety.

[0002] The present invention relates to the bunkering of marine vessels. More specifically, the present invention relates to a data-driven bunker planner system that forecasts fuel prices and optimizes bunkering costs by optimizing data related to planned routes, vessel fuel consumption, vessel fuel storage constraints, and bunkering facilities at ports of call. [Background technology]

[0003] Ship bunkering refers to the process by which a ship is supplied with fuel to continue its intended function. The bunkering of a marine vessel is called bunkering. Such a vessel can only store a finite amount of fuel on board and will therefore need to plan and visit various ports during its journey for bunkering.

[0004] Since not every port offers bunkering facilities, the ship's crew will need to plan for fuel procurement prior to or during the ship's journey to ensure that the ship will have enough fuel to complete its journey. Preferably, to minimize the ship's operating costs, the crew should consider fuel prices when bunkering at the selected bunkering port or when entering into a bunkering contract with a bunkering supplier.

[0005] However, fuel prices may vary at each bunkering port. In addition, fuel prices may vary daily during the vessel's journey. Therefore, it may be difficult for the vessel's crew to optimally plan the vessel's bunkering operations. It is desirable that alternative routes be considered or bunkering contracts be nominated for execution so as to optimize the cost of the vessel's bunkering operations, depending on the fuel price of the day.

[0006] There are many known systems that may be involved in managing ship bunkering costs. For example, U.S. Patent No. 10,139,235, assigned to Nippon Yusen Kabushiki Kaisha, describes a data processing device that generates indicative data of prescribed bunker oil prices together with rules for determining bunkering quantities at one or more bunkering locations, and outputs the data to a terminal device that displays information used to formulate a bunkering plan. The data processes include route identifiers, reference bunker oil price data, bunker oil price data, price determination rules, bunkering plan data, bunkering locations, remaining oil data, bunkering quantities, route-specific formulation rules, and port-specific formulation rules. This data processing device simplifies the complex task of formulating a ship bunkering plan.

[0007] Another known technique is in US Patent Application Publication No. 20160265920, assigned to Nippon Yusen Kabushiki Kaisha, which provides a bunkering plan to reduce fuel costs in a ship's voyage. The same application also provides a device including an acquisition unit for fuel consumption, port / port distance, voyage plan, remaining fuel, fuel price, bunkering plan, fuel cost calculation, and output notification related to fuel cost to provide a bunkering plan.

[0008] However, it should be noted that US Patent No. 10,139,235 does not guarantee the quality and performance of predefined decision rules in terms of bunkering cost optimization. Meanwhile, US Patent Application No. 20160265920 generates only one feasible bunkering plan instead of an optimal bunkering plan that minimizes the expected future fuel cost. Therefore, it would be desirable to have an invention that is data-driven to optimize bunkering costs as the ship travels from one port to another according to a planned route. Preferably, some data relates to forecasting fuel prices and / or fuel indices and determining the optimal time, location and amount to fulfill the bunkering contract based on the forecasted fuel prices or fuel indices. Summary of the Invention [Means for solving the problem]

[0009] The present invention provides a data-driven bunker planner system for optimizing bunkering costs for marine vessels. To achieve this objective, the present invention provides a computing system and a bunker planning system on board a vessel. Specifically, the bunker planning system further includes a fuel price prediction module and a data optimization module. The prediction module will predict future fuel prices and / or fuel indices in the long or short term. The data optimization module not only determines an optimal route based on the predicted fuel prices and / or fuel indices, but also determines an optimal time to fulfill one or more bunkering period contracts based on the predicted fuel prices and / or fuel indices.

[0010] Advantageously, the present invention supports decision making in dynamic settings by using data-driven non-parametric techniques. Furthermore, the use of forecasting and data optimization allows decision makers to determine the most economical procurement quantities of fuel and create appropriate fuel procurement plans that induce the lowest costs. Moreover, the data-driven bunker planning system is compatible with or integrated with today's ship computing systems.

[0011] In one embodiment, the present invention provides a data driven bunker planning system for minimizing ship bunkering costs. The data driven bunker planning system forecasts fuel prices and / or fuel indices and optimizes data related to planned routes, ship fuel consumption, ship constraints, and ports of call.

[0012] Preferably, the bunker planning system further determines the optimal time, amount and location to refuel to fulfill one or more bunkering term contracts based on forecasted fuel prices and / or fuel indexes.

[0013] Preferably, the bunker planning system further includes a data acquisition module.

[0014] Preferably, the data acquisition module further includes a bunkering term contract data acquisition sub-module, a characteristics data acquisition sub-module, a fuel data acquisition sub-module, a route data acquisition sub-module, a fuel consumption data acquisition sub-module, a vessel status data acquisition sub-module and a port data acquisition sub-module.

[0015] Preferably, the bunker planning system further includes a fuel price forecast module.

[0016] Preferably, the fuel price forecasting module further includes a long-term forecasting sub-module and a short-term forecasting sub-module.

[0017] Preferably, the bunker planning system further includes a data optimization module.

[0018] Preferably, the optimization module further includes a bunker plan optimization sub-module and a nomination date optimization sub-module.

[0019] Preferably, the bunker planning system further includes a notice generation module.

[0020] Preferably, the notification generation module generates any one of a combination of notifications including forecasted fuel prices and / or fuel indices, bunkering contract designations, and optimized bunkering plans.

[0021] Preferably, the fuel price prediction module receives data from the characteristic data acquisition sub-module and the fuel price data acquisition sub-module.

[0022] Preferably, the fuel price forecasting module employs customized machine learning and optimization models for long-term and short-term forecasting selected from the following: Ensemble Lasso, Random Forest, XGBoost, Support Vector Machine, trained Structural Prescriptive-Empirical Risk Management (SP-ERM) and trained Truncated Scenario-wise Linear Decision Rule-Empirical Risk Management (TSLDR-ERM).

[0023] Preferably, the data optimization module receives data from the data acquisition module and the fuel price forecasting module.

[0024] Preferably, the bunker planning system is in communication with a server to feed information to the data acquisition module.

[0025] Preferably, the bunker planning system further includes a human-machine interface system to enable a decision maker to input information and for the bunker planning system to output an optimized bunkering plan.

[0026] In another embodiment, a method for minimizing bunkering costs of a ship is provided, the method comprising forecasting fuel prices and / or fuel indices by using a data-driven bunker planning system on the ship. Preferably, the method further comprises determining an optimal time, amount and location to bunker to fulfill one or more bunkering period contracts based on short-term, long-term or combined short-term and long-term forecasted fuel prices and / or fuel indices. Preferably, the step of forecasting fuel prices and / or fuel indices employs customized machine learning and optimization models for long-term and short-term forecasting selected from the following: Ensemble Lasso, Random Forest, XGBoost, Support Vector Machine, trained Structural Prescriptive-Empirical Risk Management (SP-ERM) and trained Truncated Scenario-wise Linear Decision Rule-Empirical Risk Management (TSLDR-ERM).

[0027] Those skilled in the art will readily appreciate that the present invention is well adapted to carry out the objects and obtain the ends and advantages mentioned as well as the advantages inherent therein. The embodiments described herein are not intended as limitations on the scope of the present invention.

[0028] In order to facilitate an understanding of the invention, there are illustrated in the accompanying drawings preferred embodiments, from a consideration of which, when considered in conjunction with the following description, the invention, its structure and operation, and many of its advantages will be readily understood and appreciated. [Brief description of the drawings]

[0029] [Figure 1] 1 illustrates the overall infrastructure upon which the system and method of the present invention will be implemented. [Diagram 2]1 shows a detailed block diagram of a bunker planning system within a computing system in its preferred embodiment. [Diagram 3] 1 illustrates an example of information organized within a bunkering period contract data acquisition sub-module that is part of a data acquisition module of a bunker planning system. [Figure 4] 1 shows an example of information organized within a feature data acquisition sub-module that is part of a data acquisition module. [Diagram 5] 1 illustrates an example of information organized within a fuel price index data acquisition sub-module that is part of a data acquisition module. [Figure 6] 1 shows an example of information organized within a Route Data Acquisition sub-module that is part of the Data Acquisition module. [Figure 7] 1 illustrates an example of information organized within a fuel consumption data acquisition sub-module that is part of a data acquisition module. [Figure 8] 1 shows an example of information organized within a vessel status data acquisition sub-module that is part of the data acquisition module. [Figure 9] 1 illustrates an example of information organized within a Port Data Acquisition sub-module that is part of the Data Acquisition module. [Figure 10] 1 is a flow chart illustrating the steps performed by the bunker planning optimization sub-module that is part of the optimization module of the bunker planning system. [Figure 11] 13 is a flow chart illustrating the steps performed by the appointment date optimization sub-module that is part of the optimization module. [Figure 12] 13 illustrates an example of a first output of a notification generation module (forecasted fuel price and / or fuel index). [Figure 13] 13 illustrates an example of a second output of the notification generation module (bunker contract appointment date). [Figure 14] 13 shows an example of a third output of the notification generation module (Bunkering Plan). DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0030] The present invention relates to a data-driven bunker planner system for optimizing bunkering costs for marine vessels. The invention may also be presented in many different embodiments having common elements. According to one concept of the invention, a data-driven bunker system is placed on board a vessel and allows predicting fuel prices and / or fuel indices as the vessel proceeds on a planned route and determining a bunkering plan within a data optimization scope related to bunkering constraints, fuel consumption and bunkering facilities at ports of call.

[0031] It should be noted that in the context of the present invention the term "marine" may refer to a boat, a container ship, a cargo ship, a bulk carrier, etc.

[0032] It should be noted that in the context of the present invention, the term "port" may refer to a seaport, a bay where an ocean-based vessel may dock, or any location where an ocean-based vessel may dock, for example to load or unload cargo.

[0033] Additionally, it should be noted that a port that has a fuel supplier offering bunkering services may be referred to as a "bunkering port" or "bunkering port."

[0034] Additionally, it should be noted that ports that are planned stops for a vessel along its route may be referred to as "ports of call."

[0035] It should further be noted that in the context of the present invention, the term "forecast window" may refer to the days of the future fuel price or index to be predicted relative to when the forecast is made. The term "short term" may refer to a forecast window of 5 working days or less, and the term "long term" may refer to a forecast window of more than 5 working days.

[0036] The invention will now be described in more detail, by way of example, with reference to the accompanying drawings, in which: For ease of reference, common reference numerals or series of numerals will be used throughout the accompanying drawings when referring to the same or similar features common to the accompanying drawings, in which:

[0037] Figure 1 illustrates a diagram of the overall infrastructure on which the present invention is implemented. In particular, Figure 1 illustrates a ship 1 traveling along a pre-planned route (drawn in bold) after leaving an origin port A to arrive at a destination port Z. The ship 1 with a computing system 10 is intended to support a decision maker of the ship 1 (e.g. the ship's captain) throughout the journey of the ship 1 along the pre-planned route.

[0038] Referring to Figure 1, along the pre-planned route between origin port A and destination port Z, there are many ports located in between. Some of these ports are considered as ports of call along the pre-planned route (such as ports D, E and F). Furthermore, there may be other ports (ports P, Q and R) that provide bunkering services located near the aforementioned ports of call D, E and F.

[0039] Referring to Figure 1, based on the example itinerary of the pre-planned route, vessel 1 will stop at all ports D, E, F. However, it should be noted that not all of ports D, E, F may provide bunkering operations. For example, port D may be a port of call where vessel 1 needs to stop to unload cargo and bunker, port E may be a port of call where vessel 1 needs to stop to bunker, and port F may be a port of call where vessel 1 needs to stop to unload cargo but cannot receive bunkering because bunkering services are not provided in this port.

[0040] With these aforementioned considerations, prior to the journey of vessel 1, the decision maker may enter into a bunkering contract with the bunker supplier at port D. In the maritime context, a bunkering contract may also be called a bunkering contract. A bunkering contract is a type of contract where the decision maker agrees to a contractual obligation with the bunker supplier to buy fuel from or sell fuel to the bunker supplier at an agreed price. Furthermore, a bunkering contract may be price-based on the date of nomination (DON) or date of delivery (DOD). For a bunkering contract based on the date of nomination (DON), the unit price is determined by the fuel price on the date the bunker fuel is nominated (i.e., ordered). For example, the decision maker is allowed to place an order five days before the actual delivery date, and therefore the decision maker needs to forecast the fuel price or fuel index for the next five days. For a bunkering contract based on the date of delivery (DOD), the unit price is equal to the fuel price on the date the bunker fuel is delivered to the vessel.

[0041] 1, it is assumed that as the vessel 1 proceeds along a pre-planned route, its computing system 10 obtains status information of the vessel 1 and information from the outside world. For example, in a known manner, the vessel 1 establishes a multi-way communication link with a server 2 via a satellite 31. The server 2 is preferably at a fixed location. As the vessel 1 is sailing on the high seas, the server 2 may be on land or on an anchored buoy.

[0042] In a first use case, the server 2 may act as a data aggregator for the ship 1 by aggregating information from the outside world. The server 2 may communicate with other servers 3 via wired or wireless means to aggregate data such as fuel prices, fuel indices, port information, etc. The server 2 may then pass the aggregated data to the ship 1 via satellite 31. The computing system 10 of the ship 1 shall then process the status information of the ship 1 and the aggregated data from the outside world in order to predict fuel prices and / or fuel indices and optimize bunkering costs as the ship 1 proceeds along a pre-planned route.

[0043] In a second use case, the server 2 may act as a communications intermediary with other servers 3 so that decision makers on the vessel 1 may communicate with fuel suppliers to execute fuel buy orders or fuel sell orders based on bunkering term contracts.

[0044] In a third use case, the server 3 may act as a cloud database for storing information related to the vessel 1. This may include the vessel's itinerary, current location, parameters, crew data, etc.

[0045] Additionally, the vessel 1 may include a communications link with a weather satellite 32 for receiving information regarding marine weather conditions along its journey route.

[0046] Depending on the status information of the vessel 1 and the aggregated data from the outside world, the computing system 10 may determine an optimal route, which may be the same as the pre-planned route, or may be an alternative route that takes into account at least the vessel 1 itinerary, the agreed bunkering period contract, or the predicted fuel prices and / or fuel indices at the bunkering ports.

[0047] Based on the example itinerary, vessel 1 stops at ports D and F to unload cargo, and bunkers at port D to comply with the bunkering period contract. This means that there is flexibility in stopping at port E for bunkering. This allows the computing system 10 to determine the example optimal route (depicted by the dotted line) as seen in Figure 1. In the example optimal route, the route taken by vessel 1 has changed from port D → port E → port F to port D → port Q → port F.

[0048] To determine the example optimal route, the computing system 10 needs to consider many factors. The first factor may be the estimated time of arrival (TOA) at both ports from the current location of the vessel 1. The second factor may be whether the predicted price of fuel at port E (within the estimated TOA of the vessel) will be higher compared to the predicted price of fuel at port Q (within the estimated TOA of the vessel). In this example optimal route, the computing system 10 may determine that the estimated TOA of the vessel at both ports is more or less the same, and the predicted price of fuel at port E is more expensive compared to the predicted price of fuel at port Q. Thus, port E is dropped from the itinerary and changed to port Q in the example optimal route. Thus, the computing system 10 will determine a route that optimizes the fuel cost of the vessel 1.

[0049] To further optimize the fuel costs of the vessel 1, the computing system 10 will review the fuel term contracts related to the agreed fuel bunkering of the vessel 1 (e.g. on a daily basis). Preferably, each day, the computing system 10 will determine whether it is worth fulfilling the fuel term contracts. This is done by comparing the forecasted fuel price and / or fuel index with the current day fuel price and / or fuel index. Upon reviewing the fuel term contracts, if the computing system 10 determines that fulfilling the fuel term contracts on the current day will provide a benefit to the decision maker or owner of the vessel 1, the computing system 10 will nominate the fuel term contracts to be fulfilled. The decision maker may be informed of this via the human-machine interface (HMI) 18 of the vessel 1. Based on the example itinerary, this is preferably done on a daily basis before the vessel 1 arrives at port D.

[0050] It should be noted that the computing system 10 of the vessel 1 is a system of systems and further includes a combination of one or more mechanical, electrical, or electronic systems. Each of these systems are substantially interfaced with each other such that communication between them occurs. Furthermore, each of these systems has its own hardware and software components therein. Preferably, the computing system 10 further includes at least a data driven bunker planning system 11.

[0051] In addition, the computing system 10 may further include a fuel management system, a propulsion system, a ballast system, a cargo hold management system, a navigation system, a power supply system, a communication system, a data storage system, a positioning system, and a human-machine interface (HMI) system 18. These aforementioned systems may be implemented on board the vessel 1 by known means and therefore further details regarding these aforementioned systems will be kept to a minimum.

[0052] FIG. 2 shows a detailed block diagram of the data-driven bunker planning system 11 in the computing system 10 in its preferred embodiment. As shown, the data-driven bunker planning system 11 is a system in the computing system 10. The computing system 10 may be a conventional computer. The data-driven bunker planner system 11 may include a processor 11a that executes application software 11b, and at least one memory unit 11c. The data-driven bunker planning system 11 is a system in the computing system 10 that forecasts fuel prices and / or fuel indexes as the vessel 1 proceeds along its route and optimizes bunkering costs. To do so, the data-driven bunker planning system 11 is interfaced with substantially all of the other systems of the vessel 1.

[0053] It should be noted that the data-driven bunker planning system 11 may be implemented by other means. Because the data-driven bunker planning system 11 minimally requires only a processor 11a and a memory unit 11c, the data-driven bunker planning system 11 may be implemented with or within other systems, as well as by sharing hardware components on which the application software 11b may run. For example, the data-driven bunker planning system 11 may share the same hardware components as the human-machine interface (HMI) system 18.

[0054] In the data-driven bunker planning system 11, the processor 11a schedules and executes software instructions or computer logic instructions as instructed by the application software 11b running thereon. Preferably, the data-driven bunker planning system 11 is a conventional processor, an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or a combination thereof. Regarding the memory unit 11c, the memory unit 11c is preferably interfaced with the processor 11a. The memory unit 11c may store program files related to the application software 11b, or may temporarily store processing data during the runtime of the application software 11b.

[0055] It is further shown in FIG. 2 that the processor 11a operates one or more modules. These modules are software modules of application software 11b that operate or run on the processor 11a to calculate algorithms and generate control instructions related to the optimization of bunkering costs. These modules include a data acquisition module 111, a fuel price prediction module 112, a data optimization module 113, and a notification generation module 114. It should be noted that the presented modules do not necessarily have to be modules in a software embodiment and can be in a hardware embodiment, where the presented modules can be connected to the processor 11a or can be their own independent computer system. Additional modules can be included to provide support for the aforementioned modules.

[0056] In one embodiment, the data acquisition module 111 acquires status information of the vessel 1 and aggregated data from the outside world. Preferably, the data acquisition module 111 further includes a bunkering period contract data acquisition submodule 1111, a characteristics data acquisition submodule 1112, a fuel price data acquisition submodule 1113, a route data acquisition submodule 1114, a fuel consumption data acquisition submodule 1115, a vessel status data acquisition submodule 1116 and a port data acquisition submodule 1117.

[0057] With regard to the bunkering term contract data acquisition sub-module 1111, it preferably acquires one or more bunkering term contracts agreed between the decision maker and the fuel supplier prior to or during the journey. Preferably, the bunkering term contract data is keyed in by the decision maker through the HMI system 18 of the vessel 1. Alternatively, the bunkering term contract data will be stored in and retrieved from the server 2 or the vessel's data storage system. The bunkering term contract data includes: (i) The bunkering port or bunkering port where the bunkering contract is concluded; (ii) corresponding fuel suppliers in those bunkering ports with which term bunkering contracts have been signed; (iii) The agreed minimum quantity of fuel to be procured. (iv) The agreed maximum quantity of fuel to be procured. (v) Service costs charged by the supplier (which are fixed) (vi) Agreed fuel index to be adhered to. (vii) Insurance costs for bunkering contracts (viii) Agreed price basis for refueling period contracts, which may be date of designation (DON) or date of delivery (DOD). (ix) the estimated number of days that the bunkering contract will be executed prior to the estimated arrival of the vessel at the port; (x) the estimated number of days that the bunkering contract will expire prior to the estimated arrival of the vessel at the port;

[0058] These data may be organized by the refueling contract data acquisition submodule 1111 into the form shown in FIG.

[0059] With regard to the feature data acquisition sub-module 1112, it preferably acquires and / or derives one or more feature item data from one or more datasets that are believed to have a significant impact on future fuel prices and / or fuel indexes. These datasets are preferably aggregated by the server 2 from the outside world. In one embodiment, the feature data acquisition sub-module 1112 uses known statistical techniques to derive feature item data from the datasets. In one embodiment, the feature data includes: (i) The date on which the characteristic data is or was calculated. (ii) a first characteristic data, which may be a numerical weighting corresponding to the impact of West Texas Intermediate (WTI) crude oil prices on fuel prices; (iii) a second characteristic data, which may be a numerical weighting corresponding to the impact of the New York Stock Exchange (NYSE) Composite on fuel prices; (iv) a third characteristic data, which may be a numerical weighting corresponding to the impact of the West Texas Intermediate (WTI) crude oil price over the past three days on fuel prices; (v) a fourth data feature, which may be a numerical weighting corresponding to the impact of the previous three-day New York Stock Exchange (NYSE) Composite on fuel prices;

[0060] These data can be organized by the characteristic data acquisition submodule 1112 into the form shown in FIG.

[0061] It should be noted that some feature item data may be considered as "long-term prediction window" data, and some feature item data may be considered as "short-term prediction window" data. For example, the first and second feature item data are considered as "long-term prediction window" data because they do not have a specific range in the dataset from which they are derived. On the other hand, the third and fourth feature item data may be considered as "short-term prediction window" data because the dataset from which they are derived has a specific range over a short period of time (in this example, the past three days).

[0062] Furthermore, the number of feature data is not limited to the number mentioned above. The number of feature data may be any number that is indicative of the impact of the data set on future fuel prices and / or fuel indices. The feature data may include the impact of the depot's fuel refining capacity on future fuel prices and / or fuel indices, the impact of the depot's fuel refining output on future fuel prices and / or fuel indices, and the like.

[0063] With regard to the fuel price data acquisition submodule 1113, it is preferable to acquire one or more fuel prices and / or fuel indices. These fuel prices and / or fuel indices are preferably aggregated by the server 2 from the outside world. If necessary, historical data of fuel prices and / or fuel indices may also be acquired. For example, the fuel price data includes: (i) The date of the fuel price or fuel index; (ii) a first fuel index, which may be the Bunker World Index (BWI) value; (iii) a second fuel index, which may be the Bunker World 380 Index (BW380) value; (iv) a third fuel index, which could be a figure for Average Bunker Fuel Prices in Asia-Pacific (APAC);

[0064] These data may be organized by the obtain fuel price data submodule 1113 into the form shown in Figure 5. Furthermore, the number of fuel prices and / or fuel indexes may not be limited to the number mentioned above and may be any number.

[0065] With regard to the route data acquisition sub-module 1114, route information of the vessel 1 is acquired, preferably based on the vessel's itinerary. For example, the route information includes origin ports, destination ports, stop ports where cargo will be loaded or unloaded, and stop ports where bunkering is allowed. In addition to this, the arrival and departure dates of the vessel 1 are included as well. These data may be keyed in by a decision maker into the HMI 18 interface of the vessel 1, or may be acquired from the server 2 or the vessel's data storage system. These data may be organized by the route data acquisition sub-module 1114 into a form as shown in FIG. 6.

[0066] With regard to the fuel consumption data acquisition sub-module 1115, it is preferable to acquire fuel consumption data of the vessel 1 based on information acquired from other systems in the computing system 10. For example, the fuel consumption data includes: (i) Port-to-port details (ii) The estimated distance between one port and another port. (iii) The estimated speed of the vessel when traveling between one port and another port. (iv) The estimated fuel consumption of the vessel when traveling between one port and another These data may be organized by the fuel consumption data acquisition submodule 1115 into the form shown in FIG.

[0067] With regard to the vessel status data acquisition sub-module 1116, it preferably acquires the current status of the vessel at any point during the vessel's voyage based on information acquired from other systems within the computing system 10. For example, the vessel status data may include: (i) Name of the ship (ii) The ship's International Maritime Organization (IMO) number. (iii) The intended service of the vessel (iv) Time stamp of the state update (v) Planned speed of the vessel (vi) Actual speed of the vessel (vii) Forecasted weather conditions along the current route (which may further include wind speed, sea temperature, etc.) (viii) The quantity of cargo present in the holds of the ship; (ix) Remaining fuel amount on board (ROB:remaining on board) (x) The amount of fuel consumed (xi) Estimated distance from the current location to another port These data may be organized by the fuel consumption data acquisition submodule 1116 into the form shown in FIG.

[0068] With regard to the port data acquisition submodule 1117, data is acquired regarding ports providing refueling services, preferably located along or near the pre-planned route. These data are preferably aggregated by the server 2 from the outside world. For example, the port data includes: (i) Name of port terminal (ii) Port coordinates (iii) Fuel types offered by the port (Very Low Sulphur Fuel Oil (VLSFO), Heavy Fuel Oil (HFO) and Ultra Low Sulphur Fuel Oil (ULSFO)); (iv) Port Draft Limits These data may be organized by the port data acquisition submodule 1117 into a format as shown in FIG.

[0069] In one embodiment, the bunker price prediction module 112 predicts the fuel price based on the data provided by both the feature data acquisition submodule 1112 and the fuel price acquisition submodule 1113. Preferably, the bunker price prediction module 112 further includes a long-term price prediction submodule 1121 and a short-term price prediction submodule 1122. In one embodiment, these fuel price prediction submodules 1121 and 1122 employ various customized machine learning and optimization models for long-term and short-term prediction, including Ensemble Lasso, Random Forest, XGBoost, Support Vector Machine, trained Structural Prescriptive-Empirical Risk Management (SP-ERM) and / or trained Truncated Scenario-wise Linear Decision Rule- Empirical Risk Management (TSLDR-ERM). Depending on the prediction window (or the period (days) of the future fuel price or index to be predicted relative to when the prediction is made), the best performing model is selected and integrated into the prediction submodules 1121 and 1122. Preferably, the forecasting method includes (i) long term decision making and (ii) short term decision making. The long term decision making is based on long term forecast fuel prices and / or fuel indexes, e.g., with a forecast window longer than 5 days, and determines the optimal combination of contracts given a portfolio of contracts. The short term decision making is based on short term forecasts, e.g., with a forecast window of 5 days or less, and determines the optimal time to place an order given the time periods and contracts defined in the bunkering period contract data in sub-module 1111.

[0070] Regarding the long-term forecasting submodule 1121, preferably, feature data (such as the above-mentioned first feature item data and second feature item data) considered as data over a long-term forecast window period is used to generate a numerical value representing a fuel price and / or a fuel index over a long-term forecast window. In another embodiment, long-term forecasting may refer to a forecast window longer than 5 days. In addition, the planning period for long-term decision-making may be substantially one month, meaning that a decision maker needs to make a refueling decision (when, where, and how much to buy) for the next month.

[0071] Regarding the short-term prediction submodule 1122, preferably, the feature data (such as the third feature item data and the fourth feature item data described above) considered as data over a short-term prediction window is used to generate a numerical value representing the fuel price and / or fuel index within the short-term prediction window. In another embodiment, short-term prediction may refer to a prediction window of 5 days or less, meaning that a decision maker needs to predict the fuel price for the next 5 days, and is allowed to place an order for refueling 5 days before the actual delivery date. Short-term prediction requires higher accuracy, but is relatively easy to determine. Long-term prediction is more challenging and may employ various customized machine learning and optimization models to predict over a longer window period.

[0072] In one embodiment, the data optimization module 113 optimizes bunkering costs as the vessel 1 travels along its journey. Preferably, the data optimization module 113 further includes a bunker plan optimization sub-module 1131 and a nomination date optimization sub-module 1132.

[0073] Specifically, the bunkering plan optimization submodule 1131 will determine the time, amount, and location for bunkering where bunkering costs can be minimized, while the nomination date optimization submodule 1132 will nominate the optimal time to place an order given a bunkering period contract. Both sets of information enable the decision maker to operate the vessel 1 along a planned route with minimal bunkering costs.

[0074] Figure 10 shows the A-series operational flow performed by the bunker plan optimization submodule 1131. The bunker plan optimization submodule 1131 will use information from the data acquisition module 111 and the forecasted long-term fuel prices and / or fuel indices from the long-term forecast submodule 1121 of the fuel price forecast module 112 to perform a series of steps to determine the time, amount and location for bunkering where bunkering costs can be minimized.

[0075] First, in step A1, a decision maker on vessel 1 interacts with HMI system 18 on vessel 1 to define a planning horizon spanning a future period for bunker planning optimization submodule 1131. The planning horizon can be a customizable time frame, for example, a 90 day time frame from the current location and time, a 180 day time frame after vessel 1 leaves port, a time frame where the vessel at least visits at least two future ports in the itinerary, etc.

[0076] Next, in step A2, the bunker plan optimization submodule 1131 simulates the future behavior of the vessel within the defined planning period. For example, the bunker plan optimization submodule 1131 simulates the port arrival time of the vessel 1 based on the route data acquired from the route data acquisition submodule 1114 and the current status of the vessel 1 acquired from the vessel status data acquisition submodule 1116.

[0077] Next, in step A3, the bunker plan optimization submodule 1131 calculates the fuel price per unit per port visit according to the simulation information from step A2, the bunkering period contract information and the forecasted long-term fuel price and / or fuel index.

[0078] Next, in step A4, the bunker plan optimization sub-module 1131 obtains information regarding the current amount of remaining fuel on board (ROB) of the vessel 1 from the vessel status data acquisition sub-module 1116.

[0079] Next, in step A5, the bunker plan optimization submodule 1131 defines one or more bunkering constraints that may affect the amount of fuel ROB in the future (e.g., within the time of the defined planning horizon). For example, these bunkering constraints may include: (i) Fuel tank capacity (ii) Utilization of different types of fuel among fuel tanks (iii) Future weather conditions for the route (iv) The minimum and maximum quantities of fuel to be procured as specified in the bunkering contract. (v) Vessel 1's speed (vi) Cargo remaining in the cargo hold area of ​​Vessel 1 (vii) Electricity usage of vessel 1

[0080] Then, in step A6, the bunker plan optimization submodule 1131 attempts to minimize the bunkering costs within a defined planning horizon based on these bunkering constraints.

[0081] Then, in final step A7, the bunker plan optimization submodule 1131 generates a bunkering plan within the defined planning horizon and refueling constraints.

[0082] Referring to FIG. 1 as well as the above description thereof, an exemplary optimal route may be generated through the aforementioned series of steps performed by the bunker plan optimization submodule 1131.

[0083] 11 illustrates the B-series operational flow performed by the appointment date optimization submodule 1132. The appointment date optimization submodule 1132 will perform a series of steps that enable decision makers to fulfill their contractual obligations at the optimal time based on data from the bunkering period contract acquisition submodule 1111 and the forecasted short-term fuel prices and / or fuel indices from the short-term price forecast submodule 1122 of the fuel price forecast module 112.

[0084] In step B1, the appointment date optimization submodule 1132 generates an ordered list of ports for which bunkering period contracts have been signed and sorted by vessel 1's scheduled arrival time.

[0085] Next, in step B2, the appointment date optimization submodule 1132 indexes each port in the list as a task i that needs to be performed, where i is defined as an ordered sequence of 1, 2, ..., N, where N is the number of ports in the list.

[0086] Next, in step B3, the appointment date optimization submodule 1132 selects the tasks to be performed.

[0087] Following step B3 is step B4. Step B4 is a decision step, where it is determined whether all tasks have been marked as "Done". If not, step B4 proceeds to step B5. If so, step B4 proceeds to step B15.

[0088] In step B5, the nomination date optimization submodule 1132 obtains further details of the task (such as the corresponding bunkering period contract for the port and the estimated arrival time of vessel 1 at the port).

[0089] Following from step B5 is step B6. Step B6 is another decision step in which it is determined whether the bunkering period contract for the task has an agreed upon price basis that is the "Date of Nomination (DON)". If so, step B6 proceeds to step B7. If not, step B6 proceeds to step B16.

[0090] Step B7 is another decision step where it is determined whether the "date of appointment" price-based bunkering period contract has entered the appointment period based on that date. If so, step B7 proceeds to step B8. If not, step B7 proceeds to step B17.

[0091] Step B8 is another decision step in which it is determined whether the "date of appointment" price-based bunkering period contract has reached the deadline for the appointment period based on the current date. If not, step B8 proceeds to step B9. If so, step B8 proceeds to step B18.

[0092] In step B9, the nomination date optimization submodule 1132 takes the forecasted short-term fuel price and / or fuel index and fits it into the current window of the nomination period, which is defined to be the duration between the current day and the closing date of the nomination period.

[0093] Next, in step B10, the nomination day optimization submodule 1132 compares the forecasted short-term fuel prices and / or fuel indices with the current day's fuel prices and / or fuel indices.

[0094] The next step from step B10 is step B11. Step B11 is another decision step in which it is determined whether the current day fuel price and / or fuel index is lower than the forecasted short-term fuel price and / or fuel index. If not, step B11 proceeds to step B12. If so, step B11 proceeds to step B13.

[0095] In step B12, because it is determined that the current day fuel price and / or fuel index is more expensive than the forecasted short-term fuel price and / or fuel index, the task will be marked as "Not Done" and will be retained as a task that needs to be re-elected at a later time.

[0096] In step B13, the current day is deemed to be the optimal time to fulfill the refueling period contract because it has been determined that the fuel price and / or fuel index for the current day is lower than the forecasted short-term fuel price and / or fuel index forecasted by the price forecast module 112. Accordingly, a task is dispatched by the appointment date optimization submodule 1132 for the refueling period contract to be fulfilled on the current day.

[0097] Following from step B13 is step B14, in which the task will be marked as "completed" and its corresponding index will be ignored from now on by the appointment date optimization submodule 1132.

[0098] Now steps B16, B17 and B18 will be described.

[0099] In step B16, it was determined that the bunkering period contract for the task does not have an agreed upon price basis that is “Date of Designation (DON)”, so the Date of Designation Optimization submodule 1132 interprets that the task will be ignored entirely.

[0100] Thus, the next step from step B16 is step B14, which causes the task to be marked as "completed" and its corresponding index to be ignored by the appointment date optimization submodule 1132 from this point on.

[0101] In step B17, since it is determined that the refueling period contract, which is the price basis for the "designated date", is not yet in the designated period based on the current date, the designated date optimization submodule 1132 interprets that the task will be temporarily ignored.

[0102] Thus, following step B16 is step B12, which causes the task to be marked as "not completed" and to be retained as a task that needs to be re-elected at a later time.

[0103] In step B18, since it is determined that the deadline for the appointment period for the “appointment date” price-based refueling period contract has been reached based on the current day, the appointment date optimization submodule 1132 interprets that the task needs to be appointed immediately.

[0104] Thus, following from step B18 is step B13, whereby a task is assigned by the Assignment Date Optimization submodule 1132 for the term contract to be fulfilled on that day. Again, following from step B13 is step B14, where the task will be marked as "completed" and its corresponding index will be ignored by the Assignment Date Optimization submodule 1132 from this point on.

[0105] Following from step B12 or B14 is step B3. This means that when marking a task as "incomplete" or "completed", the appointment day optimization submodule 1132 again picks the task as the next task to be performed. Preferably, the appointment day optimization submodule 1132 sequentially picks the next task based on the task index. This can be done by incrementing the task index i by a value of 1, so that the next task should have an index of i+1. Then, step B3 goes to step B4 again, and the whole process flow starts again for this next task.

[0106] Next, step B15, which may be considered the final step in the B-series operational flow, is described. In step B15, since it is determined that all tasks have been marked as "completed," the nomination date optimization submodule 1132 interprets all tasks as being completed. This ends the B-series operational flow. The nomination date optimization submodule 1132 may then notify a decision maker through the HMI system 18 of the vessel 1 that all tasks have been completed.

[0107] The notification generation module 114 generates one or more notifications to be presented to a decision maker via the HMI system 18 of the vessel 1. For example, the notification generation module 114 shall output at least three different notifications including: (i) forecasted fuel prices and / or fuel indexes, (ii) bunkering period contract designations, and (iii) bunkering plan for the remainder of the journey.

[0108] The first output of the notification generation module 114, the forecasted fuel price and / or index, may be organized by the vessel's HMI system 18 for display to a decision maker in the manner shown in Figure 12. In one embodiment, the forecasted fuel price includes: (i) The number of days from today (ii) one or more forecast fuel prices and / or fuel indices;

[0109] For example, with respect to the number of days ahead from the current day, the number of days may vary depending on the configuration of the application software 11b. With respect to the fuel price and / or fuel index, the fuel price and / or fuel index may be from either one or both of the long-term forecast sub-module 1121 or the short-term forecast sub-module 1122 of the price forecast module 112. Preferably, the forecasted fuel price and / or index is by trimming and combining the long-term and short-term forecasted fuel price and / or index obtained from the long-term forecast sub-module 1121 and the short-term forecast sub-module 1122. For example, the fuel price and / or fuel index employed may be a midpoint value between the long-term and short-term forecasted fuel price and / or index.

[0110] For example, for days less than 40 days ahead, the forecasted short-term fuel price and / or index from the Short Term Forecast sub-module 1122 will be used, while for days more than 40 days ahead, the forecasted long-term fuel price and / or index from the Long Term Forecast sub-module 1121 will be used.

[0111] A second output of the notification generation module 114, the bunkering period contract nomination, may be organized by the vessel's HMI system 18 to be displayed to the decision maker in the manner shown in Figure 13. For example, the bunkering period contract nomination may include: (i) Task Index for the Bunkering Contract (ii) The estimated time of arrival of the vehicle at the port where the bunkering contract has been agreed. (iii) the fuel price and / or fuel index for that day; (iv) the projected fuel price and / or fuel index for the later time of arrival (TOA) of Vessel 1; (v) Indicators that a bunkering contract should be nominated for execution. (vi) Indicators that refueling contracts should remain on task.

[0112] For example, the predicted fuel price and / or fuel index for vessel 1's later TOA is from the short-term forecast sub-module 1122 of the price forecast module 112.

[0113] The third output of the notification generation module 114, the bunkering plan, may be compiled by the vessel's HMI system 18 to be displayed to a decision maker in the manner shown in Figure 14. For example, the bunkering plan is an ordered list of the status of each fuel tank in the vessel 1, and preferably includes at least the following: (i) Name of the ship (ii) The intended service of the ship (iii) Itinerary serial number (iv) Any available fuel tanks on board the ship, each identified by an index number. (v) The type of fuel acceptable for the fuel tank. (vi) Maximum capacity of fuel tanks (vii) Preferred maximum fill volume of the tank. (viii) Preferred maximum tank fill (%) (ix) The amount of dead oil in the tank (x) The ports of call that Ship 1 will visit (xi) Estimated Time of Arrival (TOA) of vessel 1 at each port (xii) The fuel consumed by the time Vessel 1 arrives at port. (xiii) Remaining fuel on board (ROB) when Vessel 1 arrives at port. (xiv) Fuel to be refilled or bunkered in tanks (xv) Remaining fuel on board (ROB) when the ship departs from the port.

[0114] Thus, the details regarding the computing system 10 and data driven bunker planner system 11 of the present invention have been fully elucidated. In addition to the advantages explained above, the present invention further provides improved ship operation management in a maritime setting. Furthermore, the present invention further provides decision makers with additional flexibility in managing their fuels as a spinoff, so that risk management strategies such as fuel price hedging can be effectively adopted.

[0115] The present disclosure includes what is described above as well as what is included within the scope of the appended claims. Although the present invention has been described in its preferred form with a certain degree of particularity, it will be understood that the present disclosure of the preferred form is made merely by way of example, and that numerous changes in the details of construction, combination and arrangement of parts may be employed without departing from the scope of the invention.

Claims

1. a bunker planning system (11) operable in a computing system (10) located on board a vessel (1); The bunker planning system (11) comprises: Operable to forecast fuel prices and / or fuel indices and to determine optimal times, amounts and locations for refueling; fulfilling one or more refueling term contracts based on said forecasted fuel prices and / or fuel indexes; Data-driven bunker planner system.

2. The bunker planning system (11) further comprises a data acquisition module (111). The system of claim 1 .

3. The data acquisition module (111) further includes a bunkering period contract data acquisition submodule (1111), a characteristic data acquisition submodule (1112), a fuel price data acquisition submodule (1113), a route data acquisition submodule (1114), a fuel consumption data acquisition submodule (1115), a vessel status data acquisition submodule (1116), and a port data acquisition submodule (1117); The system of claim 2.

4. The bunker planning system (11) further includes a fuel price forecasting module (112).

4. A system according to claim 2 or 3.

5. The fuel price forecast module (112) further includes a long-term price forecast sub-module (1121) and a short-term price forecast sub-module (1122); The system of claim 4.

6. The bunker planning system (11) further includes a data optimization module (113).

6. A system according to claim 4 or 5.

7. The data optimization module (113) further includes a bunker plan optimization submodule (1131) and a nomination date optimization submodule (1132); The system of claim 6.

8. The bunker planning system (11) further includes a notification generation module (114).

8. A system according to claim 6 or 7.

9. the notification generation module (114) generates any one of a combination of notifications including forecasted fuel prices and / or fuel indexes, bunkering period contract designations, and bunkering plans; The system of claim 8.

10. The fuel price prediction module (112) receives data from the characteristic data acquisition sub-module (1112) and the fuel price data acquisition sub-module (1113); A system according to any one of claims 4 to 9.

11. The fuel price forecasting module (112) employs various customized machine learning and optimization models for long-term and short-term forecasting selected from Ensemble Lasso, Random Forest, XGBoost, Support Vector Machine, trained Structural Prescriptive-Empirical Risk Management (SP-ERM), and trained Truncated Scenario-wise Linear Decision Rule-Empirical Risk Management (TSLDR-ERM); A system according to any one of claims 4 to 10.

12. The data optimization module (113) receives data from the data acquisition module (111) and the fuel price forecast module (112); A system according to any one of claims 6 to 11.

13. The bunker planning system (11) communicates with a server (2) to acquire data for the data acquisition sub-modules (1111-1117); A system according to any one of claims 3 to 12.

14. The bunker planning system (11) further includes a human-machine interface system (18) that enables a decision maker to input information into the bunker planning system (11) and for the bunker planning system (11) to output the bunkering plan. A system according to any one of claims 9 to 13.

15. 1. A method for optimizing bunkering costs for a vessel (1), comprising the steps of: forecasting fuel prices and / or fuel indexes by using a bunker planning system (11) located on board the ship (1), method.

16. Given a bunkering period contract, determining the optimal time to place an order in the short term or determining the optimal combination of contracts in the long term; The method of claim 15.

17. The step of forecasting fuel prices and / or fuel indexes employs various customized machine learning and optimization models for long-term and short-term forecasting selected from Ensemble Lasso, Random Forest, XGBoost, Support Vector Machine, trained Structural Prescriptive-Empirical Risk Management (SP-ERM), and trained Truncated Scenario-wise Linear Decision Rule-Empirical Risk Management (TSLDR-ERM); 17. The method according to claim 15 or 16.

18. outputting a bunkering plan based on the forecasted fuel price and / or fuel index and a result of data optimization relating to data related to planned route, fuel consumption, vessel status, and ports of call.

18. The method according to any one of claims 15 to 17.

Images