Bulk bulk cargo market ship and cargo matching method and system based on multi-objective constraint

Abstract

The invention discloses a bulk cargo market ship and cargo matching method and system based on multi-objective constraints. The method comprises the following steps: screening out characteristic attributes and data of bulk cargo ships and characteristic attributes and data of cargoes with specific correlation in different business scenes; on the basis of the ratio of the fixed ship route to all the ship routes, identifying long boats and market ships; setting a constraint function of a plurality of constraint conditions based on ship data and cargo attributes in a cargo ship scene, constructing a target function about transportation and ship renting expenditure, solving each batch of cargoes to obtain a ship set for transporting each batch of cargoes, and constructing a ship preference list of each batch of cargoes; the method comprises the following steps: setting a cargo-ship constraint function of a plurality of constraint conditions based on ship attributes and cargo data in a ship-cargo scene, constructing a target function about ship transportation fuel consumption expenditure in combination with fuel consumption, solving each ship to obtain a cargo set transported by the ship, and constructing a cargo preference list of each ship.

Description

Technical Field

[0001] The present invention relates to the technical field of ship transportation capacity scheduling and supply chain management, and particularly to a ship-cargo matching method and system for the bulk cargo market based on multi-objective constraints. Background Art

[0002] Marine ship-cargo matching is an important issue in the shipping field and is of great significance for improving voyage efficiency and reducing costs. However, current research on marine ship-cargo matching mainly focuses on stowage research for established voyage tasks, analyzing voyage revenue based on cargo weight, volume, and stowage factors after clarifying cargo screening conditions, and research on how to allocate empty containers, etc. Although these studies are somewhat helpful for ship-cargo matching, they mainly focus on single-objective optimization and lack comprehensive consideration of multi-objective constraints through the potential relationships of multiple attributes between ships and cargoes. In addition, the common practice of ship-cargo matching by shipping companies and platforms at home and abroad is that shipowners publish ship information on the platform, shippers publish cargo information on the platform, and shipowners and shippers independently screen eligible business cooperation partners through the platform and negotiate cooperation to conclude transactions. This screening and matching operation is time-consuming and laborious, and the probability of concluding a cooperation transaction is very low, making it difficult to obtain the best benefits. Summary of the Invention

[0003] To solve the problems of lack of consideration of the potential relationships of multiple attributes between ships and cargoes, long ship-cargo matching time, low transaction success rate, and poor economic benefits during ship-cargo matching in the existing bulk cargo market, the present invention provides a ship-cargo matching method for the bulk cargo market based on multi-objective constraints. This method can efficiently find the optimal ship transportation capacity with the best efficiency and benefits for shippers and the optimal loaded cargo with the best revenue for shipowners based on multi-objective constraint conditions, achieving a win-win situation for shippers and shipowners. The present invention also relates to a ship-cargo matching system for the bulk cargo market based on multi-objective constraints.

[0004] The technical solution of the present invention is as follows:

[0005] A ship-cargo matching method for the bulk cargo market based on multi-objective constraints, comprising the following steps:

[0006] Feature Attribute Selection and Data Acquisition Step: Combining the feature attributes of bulk cargo ships and goods and the actual business scenarios, screening out the feature attributes of bulk cargo ships and goods with specific relevance in different business scenarios, and acquiring the ship data and cargo data of the feature attributes. Specifically: Obtaining multiple bulk cargo ship data from the ship database, including navigation parameters and AIS data. The navigation parameters include the transportation type, deadweight tonnage, designed speed, daily rent, fuel consumption, maximum hold capacity, and draft of the ship; the AIS data includes the real-time position, MMSI, AIS information upload time, longitude and latitude information, course over ground information, origin and destination port information, and the start and end times of the voyage of the ship; obtaining multiple batches of cargo data from the cargo database, including the type, weight, volume, loading and unloading ports of the goods, and the time window including the earliest loading time and the latest unloading time.

[0007] Ship Operation Type Identification Step: Using the AIS data and navigation parameters of bulk cargo ships to determine the voyage information data of the ships, and based on the AIS data and voyage information data of bulk cargo ships, combining the route characteristics of bulk cargo ships, identifying long-term charter ships and spot market ships.

[0008] Ship Matching Cargo Step: For the ship matching cargo scenario, taking the ship data as the set of objects to be matched, using the type, time window, loading and unloading ports, volume, and weight attributes of the goods as restrictive conditions, combining the daily rent to construct an objective function for transportation charter expenditure, and based on this, setting a ship matching cargo constraint function with multiple constraint conditions including cargo type matching constraint, time window matching constraint, port matching constraint, deadweight and hold capacity constraint, draft constraint, and spot market ship constraint, solving for each batch of goods to obtain the set of ships for transporting each batch of goods, sorting them in ascending order of transportation charter cost, and constructing a ship preference list for each batch of goods to achieve matching suitable ships for the goods.

[0009] Cargo Matching Ship Step: For the cargo matching ship scenario, taking the cargo data as the set of candidate objects, using the transportation type, real-time position, transportation capacity attributes including deadweight tonnage and maximum hold capacity of the ship as restrictive conditions, combining the fuel consumption to construct an objective function for ship transportation fuel consumption expenditure, and based on this, setting a cargo matching ship constraint function with multiple constraint conditions including ship type matching constraint, time window matching constraint, empty voyage mileage constraint, transportation capacity matching constraint, and draft constraint, solving for each ship to obtain the set of goods transported by the ship, sorting them in ascending order of transportation fuel consumption expenditure, and constructing a cargo preference list for each ship to achieve matching suitable cargo for the ship.

[0010] Preferably, the method further comprises a ship-cargo bilateral matching step. For the ship-cargo bilateral matching scenario, according to the constructed ship preference list and cargo preference list, the Gale-Shapley algorithm is used to optimize the matching order of the ship and the cargo, and according to the optimized matching results, the ship-cargo bilateral matching is achieved.

[0011] Preferably, in the ship operation type identification step, the AIS data and navigation parameters of the bulk carrier are used to determine the voyage information data of the ship, including obtaining the starting port and destination port of each voyage of the ship according to the starting and ending port information in the AIS data, and determining the time range of each voyage through the voyage start and voyage end time in the AIS data, thereby determining the basic framework information of the voyage; using the real-time position, longitude and latitude information, ground heading information and AIS information upload time in the AIS data, connecting each position point in chronological order, reconstructing the navigation track of the ship, and determining the route of the ship in each voyage according to the reconstructed navigation track in combination with the starting and ending port information.

[0012] Preferably, in the ship operation type identification step, based on the AIS data and voyage information data of bulk carriers, combined with the route characteristics of bulk carriers, long-term contract ships and market ships are identified, specifically including: selecting port information data from a database, including port location, port code, port name and port group code where the port is located; matching the port information data with the extracted AIS data of bulk carriers, marking the port group code where the starting and ending ports are located for each route, and assigning a value of null if the starting and ending ports of some routes are not in the port group; counting the starting and ending port group codes in the route information, selecting the starting and ending port group codes that appear the most times, calculating the ratio of fixed routes of a certain ship that travel to and from the same starting and ending port group during the statistical period to all the routes of the ship during the statistical period, and identifying the ship as a long-term contract ship or a market ship according to the value range of the calculated ratio.

[0013] Preferably, in the cargo-to-ship matching step, a cargo-to-ship matching constraint function with multiple constraints set based on the constructed objective function about the transport charter expenditure includes a first constraint, a second constraint, a third constraint, a fourth constraint, a fifth constraint and a sixth constraint, the first constraint being a cargo type matching constraint for ships whose loaded cargo types include the type of target cargo, the second constraint being a market ship constraint excluding long-term contract ships, the third constraint being a deadweight and capacity constraint that the deadweight tonnage of the ship is greater than or equal to the weight of the cargo and the maximum capacity of the ship is greater than or equal to the volume of the cargo, the fourth constraint being a draft constraint that the draft of the ship meets the requirements of the cargo loading and unloading port, the fifth constraint being a port matching constraint that the cargo unloading port is consistent with the ship's route, and the sixth constraint being a time window matching constraint that the cargo loading and unloading time window matches the available time window of the ship.

[0014] Preferably, in the ship-cargo matching step, the ship-cargo matching constraint function set based on the constructed objective function of the ship transportation fuel consumption expenditure includes a seventh constraint, an eighth constraint, a ninth constraint, a tenth constraint and an eleventh constraint, the seventh constraint is a ship type matching constraint for selecting a cargo type belonging to the transportation type of the target ship from the candidate cargo set, the eighth constraint is a time window matching constraint for matching the cargo loading and unloading time window with the available time of the ship, the ninth constraint is a ship's empty sailing mileage constraint within an empty sailing mileage threshold range, the tenth constraint is a transport capacity matching constraint that the ship's deadweight tonnage is greater than or equal to the weight of the cargo and the ship's maximum cabin capacity is greater than or equal to the volume of the cargo, and the eleventh constraint is a draft constraint that the ship's draft meets the requirements of the cargo loading and unloading port.

[0015] Preferably, in the step of matching ships with cargoes, the objective function of transport charter expenditure constructed in combination with the daily rent includes the daily rent and the total number of days of the transport voyage, and the total number of days of the transport voyage includes the number of days for the ship to sail empty from the current position to the cargo loading port and the number of days from the ship arriving at the cargo loading port to the ship unloading the cargo.

[0016] Preferably, in the step of matching ships with cargo, the objective function of ship transportation fuel consumption expenditure constructed in combination with the fuel consumption includes the total fuel consumption of the transport voyage and the unit price of each ton of crude oil, and the total fuel consumption of the transport voyage includes the fuel consumption of the ship sailing empty from the current position to the cargo loading port and the fuel consumption of the ship sailing from the cargo loading port to the unloading port.

[0017] A ship-cargo matching system for the bulk cargo market based on multi-objective constraints, including a feature attribute selection and data acquisition module, a ship operation type identification module, a cargo-to-ship module, and a ship-to-cargo module connected in sequence; among them,

[0018] The feature attribute selection and data acquisition module is used to screen out the feature attributes of bulk cargo ships and goods with specific relevance in different business scenarios in combination with the feature attributes of bulk cargo ships and goods and the actual business scenarios, and obtain the ship data and cargo data of the feature attributes. Specifically, it obtains multiple bulk cargo ship data from the ship library, including navigation parameters and AIS data. The navigation parameters include the transportation type, deadweight tonnage, designed speed, daily rent, fuel consumption, maximum hold capacity, and draft of the ship; the AIS data includes the real-time position, MMSI, AIS information upload time, longitude and latitude information, course over ground information, origin and destination port information, and the time of voyage start and end of the ship; it obtains multiple batches of cargo data from the cargo library, including the type, weight, volume, loading and unloading ports of the goods, and the time window including the earliest loading time and the latest unloading time;

[0019] The ship operation type identification module is used to determine the voyage information data of the ship by using the AIS data and navigation parameters of the bulk cargo ship. Based on the AIS data and voyage information data of the bulk cargo ship, combined with the route characteristics of the bulk cargo ship, it identifies long-term charter ships and spot market ships;

[0020] The cargo-to-ship module is used for the cargo-to-ship scenario. Taking the ship data as the set of objects to be matched, using the type, time window, loading and unloading ports, volume, and weight attributes of the goods as restrictive conditions, and combining the daily rent to construct an objective function for the transportation charter expenditure, and based on this, setting a cargo-to-ship constraint function with multiple constraint conditions including cargo type matching constraint, time window matching constraint, port matching constraint, deadweight and hold capacity constraint, draft constraint, and spot market ship constraint. Solving for each batch of goods to obtain the set of ships for transporting each batch of goods, and sorting them in ascending order of transportation charter cost, constructing a ship preference list for each batch of goods, and realizing the matching of suitable ships for the goods;

[0021] The module for matching goods with ships is used for the scenario of matching goods with ships. Taking the goods data as the candidate object set, and using the transportation type of the ship, real-time position, and transportation capacity attributes including deadweight tonnage and maximum hold capacity as the constraint conditions, combining with the fuel consumption to construct an objective function for the fuel consumption expenditure of ship transportation. Based on this, a ship-goods matching constraint function including multiple constraint conditions such as ship type matching constraint, time window matching constraint, empty voyage mileage constraint, transportation capacity matching constraint, and draft constraint is set up. Solve for each ship to obtain the set of goods to be transported by the ship, and sort them in ascending order according to the fuel consumption expenditure of transportation, and construct the goods preference list for each ship to achieve matching suitable goods for the ship.

[0022] Preferably, the system further includes a ship-goods bilateral matching module, which is used for the ship-goods bilateral matching scenario. According to the constructed ship preference list and goods preference list, the Gale-Shapley algorithm is adopted to optimize the matching order of ships and goods, and according to the optimized matching result, the ship-goods bilateral matching is realized.

[0023] The beneficial effects of the present invention are as follows:

[0024] The present invention provides a bulk cargo market ship-cargo matching method based on multi-objective constraints. The method combines the characteristic attributes of bulk cargo ships and cargoes and actual business scenarios, screens out the characteristic attributes of bulk cargo ships and cargoes with specific correlations in different business scenarios, and obtains the ship data and cargo data of the characteristic attributes. By obtaining the bulk cargo ship data including navigation parameters and AIS data for all leases from the ship library, and obtaining the cargo data for all shipments from the cargo library, high-quality data can be provided for ship-cargo matching, thereby improving the success rate of ship-cargo matching. Specific characteristic attributes are screened for bulk cargo scenarios to ensure that the data is highly relevant to business needs, and redundant information is avoided from interfering with the matching efficiency. The real-time position and voyage information of the AIS data support dynamic updating of the ship status, so that the matching result can adapt to the real-time operation changes of the ship, and improve the timeliness and reliability of the matching. The voyage information data of the ship is determined by using the AIS data and navigation parameters of the bulk cargo ship. Based on the AIS data and voyage information data of bulk carriers, combined with the route characteristics of bulk carriers, long-term contract ships and market ships are identified. In this way, the route with the most voyages of each ship is determined according to the AIS data and navigation parameters of bulk carriers, and then based on the ratio of the number of voyages of the route to all voyages of the ship, it is judged whether the ship is a long-term contract ship or a market ship. The method is simple and efficient, thereby distinguishing different operation modes, so as to select a more appropriate matching strategy according to the operation mode of the ship. For example, market ships are more suitable for short-term and flexible matching, while long-term contract ships are more suitable for long-term and stable matching. Constraints (such as market ship constraints) can be set in a targeted manner to avoid matching failures due to ship availability issues. It can also optimize ship resource allocation and improve ship utilization for fixed routes of long-term contract ships and dynamic scheduling of market ships, reduce uncertainty in the matching process, improve matching efficiency, and ensure that ships and cargoes can quickly find the most suitable matching objects;For the scenario of cargo-to-ship, taking ship data as the set of objects to be matched, using the type of goods, time window, loading and unloading ports, volume, and weight attributes as restrictive conditions, combining with the daily rent to construct an objective function for the transportation chartering expenditure, and based on this, setting a cargo-to-ship constraint function with multiple constraint conditions including cargo type matching constraint, time window matching constraint, port matching constraint, deadweight and capacity constraint, draft constraint, and market ship constraint. Solve for each batch of goods to obtain the set of ships for transporting each batch of goods, and sort them in ascending order according to the transportation chartering cost to construct a ship preference list for each batch of goods, so as to match suitable ships for the goods. In this way, taking the multiple attributes of the goods as restrictive conditions, taking the ship data as the set of objects to be matched, and based on this, setting a cargo-to-ship constraint function with multiple constraint conditions can fully consider the multi-attribute correlation relationship between the goods and the ships, ensure that the matched ships can fully meet the transportation requirements of the goods, ensure finding the ships with the lowest cost for each batch of goods, reduce the transportation cost, and then efficiently and accurately construct a ship preference list for each batch of goods, find the ship transportation capacity with the best efficiency and benefit for the shipper, realize the optimal utilization of ship resources, and improve the operation efficiency of the ships;For the scenario of matching ships with goods, the goods data is used as the candidate object set, and the transportation type, real-time location of the ship, and transportation capacity attributes including deadweight tonnage and maximum hold capacity are used as constraint conditions. Combining the fuel consumption, a target function for the fuel consumption expenditure of ship transportation is constructed. Based on this, a ship-goods matching constraint function including multiple constraint conditions such as ship type matching constraint, time window matching constraint, empty voyage mileage constraint, transportation capacity matching constraint, and draft constraint is set. Each ship is solved to obtain the set of goods for the ship's transportation, and they are sorted in ascending order of transportation fuel consumption expenditure to construct the goods preference list for each ship, thus realizing the matching of suitable goods for the ship. In this way, by using multiple attributes of the ship as constraint conditions, the goods data as the candidate object set, and based on this, setting a ship-goods matching constraint function with multiple constraint conditions, the multi-attribute correlation relationship between the ship and the goods can be fully considered, ensuring that the goods with the lowest fuel consumption are found for each ship, reducing the operating cost, and then efficiently and accurately constructing the goods preference list for each ship, realizing the optimal allocation of goods resources, improving the transportation efficiency of goods, and finding the most profitable loaded goods for the shipowner. Through feature attribute selection and data acquisition, ship operation type identification, and in the scenario of matching goods with ships, setting a goods-ship matching constraint function with multiple constraint conditions based on goods attributes, solving and constructing the goods preference list for each ship, and in the scenario of matching ships with goods, setting a ship-goods matching constraint function with multiple constraint conditions based on ship attributes, solving and constructing the goods preference list for each ship, it is possible to efficiently find the most efficient and profitable ship capacity for the shipper and the most profitable loaded goods for the shipowner based on the multi-objective constraint conditions of the ship and the goods, achieving a win-win situation for both the shipper and the shipowner, significantly improving the matching efficiency, reducing the transportation cost, optimizing the resource utilization, improving the operation reliability, and bringing environmental benefits. This method is not only applicable to the bulk cargo market but also can be extended to other types of shipping operations, with broad application prospects and significant economic benefits.

[0025] The present invention also provides a ship-cargo bilateral matching method. For the ship-cargo bilateral matching scenario, according to the constructed ship preference list and cargo preference list, the Gale-Shapley algorithm is adopted to continuously optimize the matching to ensure the stability of the final matching result, and efficiently achieve ship-cargo bilateral matching, thereby achieving a win-win situation for the shipowner and the shipowner. The bulk cargo market ship-cargo matching method based on multi-objective constraints is for three business scenarios, namely, cargo-to-ship, ship-to-cargo and ship-to-cargo bilateral matching. It can be understood as constructing corresponding multi-objective (attribute) constraint models and objective functions respectively, dispersing various constraints in multiple stages for consideration, and better limiting the number of solutions in the solution space, so as to quickly solve the multi-objective (attribute) matching problem, and can more quickly select the most suitable ship for cargo to ship or select the most suitable cargo for the ship to occupy its transportation capacity, thereby reducing the empty sailing time, mileage and waiting time of the ship, reducing the loss caused by empty load, and controlling the backlog, detention time and cost of the cargo to be transported.

[0026] The present invention utilizes AIS data of bulk cargo ships, connects various position points in chronological order, reconstructs the ship's navigation track, and determines the ship's route in each voyage based on the reconstructed navigation track in combination with the starting and ending port information, so that the route of each voyage can be obtained accurately and reliably.

[0027] The present invention marks the port group code of the starting and ending ports for each route, counts and selects the port group code of the starting and ending ports with the most occurrences, calculates the value range of the ratio of the fixed routes of a certain ship that go back and forth with the same starting and ending port group to all the routes of the ship in the statistical time period, identifies whether the ship is a long-term contract ship or a market ship, calculates the ratio of the ship's sailing frequency on a specific route to its total routes, and can more accurately distinguish whether the ship tends to have a fixed route (long-term contract ship) or a flexibly selected route (market ship).

[0028] The present invention sets a cargo-ship matching constraint function with multiple constraints based on the constructed objective function about the transport charter expenditure. The constraints include the cargo type matching constraint of the ship whose loaded cargo type includes the type of the target cargo, the time window matching constraint that the cargo loading and unloading time window matches the available time window of the ship, the port matching constraint that the cargo loading and unloading port is consistent with the ship's route, the deadweight and capacity constraints that the ship's deadweight tonnage is greater than or equal to the weight of the cargo and the ship's maximum cargo capacity is greater than or equal to the volume of the cargo, the draft constraint that the ship's draft meets the requirements of the cargo loading and unloading port, the market ship constraint that excludes long-term contract ships, etc. The ship data is constrained from the multi-dimensional attributes of the cargo, and the ship's transportation capacity with the best efficiency and benefits can be efficiently found for the ship owner.

[0029] The present invention sets a ship-cargo constraint function with multiple constraints based on the constructed objective function about the ship transportation fuel consumption expenditure. The constraints include a ship type matching constraint that the cargo type belongs to the transportation type of the target ship from the candidate cargo set, a time window matching constraint that the cargo loading and unloading time window matches the available time of the ship, a ship's empty sailing mileage constraint that the empty sailing mileage is within the empty sailing mileage threshold range, a transport capacity matching constraint that the ship's deadweight tonnage is greater than or equal to the weight of the cargo and the ship's maximum cabin capacity is greater than or equal to the volume of the cargo, a draft constraint that the ship's draft meets the requirements of the cargo loading and unloading port, etc. The cargo data is constrained from the multi-dimensional attributes of the ship, and the shipowner can efficiently find the most profitable loaded cargo.

[0030] The objective function of transport charter expenditure constructed by the present invention includes the daily rent and the total number of days of the transport voyage, wherein the total number of days of the transport voyage includes the number of days for the ship to sail empty-handed from the current position to the cargo loading port and the number of days from the ship arriving at the cargo loading port to the ship unloading the cargo. In the cargo-ship matching scenario, the matching of port attributes adopts non-exact matching, that is, the time of the ship and the cargo coincides, but the port may have deviations. Therefore, the calculation of the total number of days of the transport voyage includes the number of days for the ship to sail empty-handed, which can make the calculation of the transport charter expenditure more accurate.

[0031] The objective function of the ship transportation fuel consumption expenditure constructed by the present invention includes the total fuel consumption of the transport voyage and the unit price of each ton of crude oil, wherein the total fuel consumption of the transport voyage includes the fuel consumption of the ship sailing empty from the current position to the cargo loading port and the fuel consumption of the ship sailing from the cargo loading port to the unloading port. In the cargo-ship matching scenario, the matching of port attributes adopts non-exact matching, that is, the time of the ship and the cargo coincides, but the port may have deviations. Therefore, the calculation of the total fuel consumption of the transport voyage includes the fuel consumption of the ship sailing empty, which can make the calculation of the ship transportation fuel consumption expenditure more accurate.

[0032] The present invention also relates to a bulk cargo market ship-cargo matching system based on multi-objective constraints, which corresponds to the above-mentioned bulk cargo market ship-cargo matching method based on multi-objective constraints, and can be understood as a system that implements the above-mentioned bulk cargo market ship-cargo matching method based on multi-objective constraints, including a feature attribute selection and data acquisition module, a ship operation type identification module, a cargo-to-ship matching module, and a ship-to-cargo matching module. Each module works in coordination with each other, and is a convenient and highly reliable bulk cargo market ship-cargo matching system that can quickly select the most suitable ship for cargo to ship or select the most suitable cargo for the ship to occupy its transportation capacity, thereby reducing the empty sailing time, mileage and waiting time of the ship, reducing the loss caused by empty load, and reducing the backlog, detention time and cost of the cargo to be transported. The ship operation type identification module can effectively identify the long-term contract ship and the market ship in the market by calculating the proportion of the ship's sailing frequency on a specific route to its total route, and conveniently select the market ship for subsequent ship-cargo matching, thereby reducing the complexity of matching. The present invention adopts multi-objective constraints. In the scenario of matching cargo to ship, the matching cargo to ship module takes multiple attributes of cargo as constraints, takes ship data as a set of objects to be matched, and sets multiple constraints on this basis. The constraint function of matching cargo to ship can fully consider the multi-attribute relationship between cargo and ship, and then efficiently and accurately construct a ship preference list for each batch of cargo, so as to find the ship capacity with the best efficiency and benefit for the ship owner; in the scenario of matching ship to cargo, the matching ship to cargo module takes multiple attributes of ship as constraints, cargo data as a set of candidate objects, and sets multiple constraints on this basis. The constraint function of matching ship to cargo can fully consider the multi-attribute relationship between ship and cargo, and then efficiently and accurately construct a cargo preference list for each ship, so as to find the loaded cargo with the best benefit for the ship owner. The matching cargo to ship module and the matching ship to cargo module automatically complete the objective function solution of multiple constraints in the processor, which reduces the analysis cost of matching cargo to ship and matching ship to cargo, improves the matching efficiency, and makes the matching of ship and cargo in the bulk cargo market more efficient and economical. Through data-driven multi-objective constraint optimization, the system achieves two-way precision, dynamic and cost optimization of ship-cargo matching in the bulk cargo market, effectively solving the problems of low efficiency and high cost of traditional matching. It can significantly improve ship-cargo matching efficiency, reduce transportation costs, optimize resource utilization, improve operational reliability, and bring environmental benefits. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 The present invention is a flow chart of a bulk cargo market ship-cargo matching method based on multi-objective constraints.

[0034] Figure 2 This is an example diagram of the route from the cargo loading port to the cargo unloading port of the present invention.

[0035] Figure 3This is the structural block diagram of the ship-cargo matching system in the bulk cargo market based on multi-objective constraints of the present invention. Detailed implementation manners

[0036] To understand the content of the present invention more clearly, it will be described in detail in conjunction with the drawings and embodiments.

[0037] The present invention discloses a ship-cargo matching method in the bulk cargo market based on multi-objective constraints. Aiming at solving problems such as the failure to consider the multi-attribute potential connection between goods and ships during ship-cargo matching, long time-consuming for manual ship-cargo matching, low transaction success rate, and poor economic benefits. When performing ship-cargo matching, based on the multi-attributes and data of ships and goods, a target function including multiple constraint conditions is set, and the cargo preference list for each ship or the ship preference list for each batch of goods is solved and constructed to efficiently and accurately achieve the optimal matching of ships and goods in the bulk cargo market. This method can be executed by a processor or a processing-capable electronic device, such as Figure 1 As shown, obtain the attributes, data of bulk cargo ships with specific relevance and the attributes, data of goods in different business scenarios. Then, based on the AIS data and voyage information data of bulk cargo ships, identify long-term charter ships and spot market ships. In the scenario of matching goods to ships, using various attributes of goods as limiting conditions and ship data as the set of objects to be matched, set a cargo-to-ship constraint function with multiple constraint conditions, and solve and construct the ship preference list for each batch of goods. In the scenario of matching ships to goods, using various attributes of ships as limiting conditions and cargo data as the set of objects to be matched, set a ship-to-cargo constraint function with multiple constraint conditions, and solve and construct the cargo preference list for each ship. The present invention adopts multi-objective constraint conditions. In the scenario of matching goods to ships, the cargo-to-ship module uses various attributes of goods as limiting conditions, ship data as the set of objects to be matched, and based on this, sets a cargo-to-ship constraint function with multiple constraint conditions, which can fully consider the multi-attribute correlation relationship between goods and ships, and then efficiently and accurately construct the ship preference list for each batch of goods to find the ship transportation capacity with the best efficiency and benefit for the cargo owner; in the scenario of matching ships to goods, the ship-to-cargo module uses various attributes of ships as limiting conditions, cargo data as the set of candidate objects, and based on this, sets a ship-to-cargo constraint function with multiple constraint conditions, which can fully consider the multi-attribute correlation relationship between ships and goods, and then efficiently and accurately construct the cargo preference list for each ship to find the most profitable loaded goods for the shipowner. Specifically, the method includes the following steps:

[0038] 1. Feature attribute selection and data acquisition steps: Combining the feature attributes of bulk cargo ships and goods, as well as the actual business scenarios, filter out the feature attributes of bulk cargo ships and goods with specific relevance in different business scenarios, and obtain the ship data and cargo data of these feature attributes. Specifically: Obtain multiple bulk cargo ship data from the ship database, including navigation parameters and AIS data. The navigation parameters include the transportation type, deadweight tonnage, designed speed, daily rent, fuel consumption, maximum hold capacity, and draft of the ship; the AIS data includes the real-time position, MMSI, AIS information upload time, longitude and latitude information, course over ground information, origin and destination ports information, and the start and end times of the voyage. Obtain multiple batches of cargo data from the cargo database, including the type, weight, volume, loading and unloading ports of the cargo, and the time window including the earliest loading time and the latest unloading time.

[0039] There are many factors related to ship-cargo matching in the voyage estimation stage, mainly including ships, goods, ports, canals, travel distance, fuel, and tariffs, etc. Among them, the ship data related to ship-cargo matching in the voyage estimation stage mainly includes the transportation type, deadweight tonnage, designed speed, fuel consumption, rent, maximum hold capacity, draft, etc. of the ship; the cargo data mainly includes the type, quantity, packaging form, weight, stowage factor (stowage factor, broken stowage factor), volume, loading and unloading ports, and the time window of the earliest loading time and the latest unloading time of the cargo, etc.; the port data mainly includes the maximum draft, cargo handling rate, port charges, port location information, port code, Chinese name of the port, and port group code where the port is located, etc.; the canal data mainly refers to the calculation method of canal charges; the travel distance refers to the travel distance of the ship in a voyage (idle running of the ship will result in dead mileage charges), including the point-to-point distance; the fuel data mainly includes the unit price of fuel, the location of the refueling port, etc.; the tariff refers to the import and export tariffs of the cargo. There are often internal connections between these factors, such as between ships and goods, ships and canals, ships and ports, goods and ports, goods and tariffs, etc. Therefore, these factors cannot be analyzed in isolation during decision-making, but must be considered comprehensively. Therefore, the embodiments of the present invention filter out the feature attributes of bulk cargo ships and goods with specific relevance in different business scenarios through analysis, and obtain the ship data and cargo data of these feature attributes, providing high-quality data for ship-cargo matching, and thus improving the success rate of ship-cargo matching. At the same time, the embodiments of the present invention obtain AIS data including the real-time position, MMSI, AIS information upload time, the upload time of the previous AIS information, longitude and latitude information, course over ground information, origin and destination ports information, and the start and end times of the voyage of the ship, etc., to identify whether the ship is a long-term charter ship or a spot market ship.

[0040] II. Steps for identifying the types of ship operations. Determine the voyage information data of the ship by using the AIS data and navigation parameters of the bulk cargo ship. Based on the AIS data and voyage information data of the bulk cargo ship, combined with the route characteristics of the bulk cargo ship, identify the long-term charter ships and spot market ships.

[0041] Long-term charter ships and spot market ships are two main forms in the shipping market. A long-term charter ship refers to a long-term charter contract signed between the shipowner and the charterer, usually involving multiple voyages within a certain period of time; a spot market ship refers to a shipowner docking the ship in the ship market and renting it to the lessee at the market price, usually a one-time contract. In practical applications, how to accurately distinguish between long-term charter ships and spot market ships is a difficult problem. Since the voyages of long-term charter ships are usually relatively fixed, while spot market ships are more flexible and can adjust voyages at any time according to market demand, the lease term of long-term charter ships is usually relatively long, generally several months or years, while spot market ships are usually leased on the basis of a single voyage. Therefore, in the embodiments of the present invention, the proportion of fixed routes within a certain time span is statistically analyzed to determine whether it is a long-term charter ship or a spot market ship. Further, the embodiments of the present invention can set the time span for identification to be 3 months, and according to the selected time interval, select the AIS data of Chinese coastal dry bulk cargo ships within this time period from the data obtained in the feature attribute selection and data acquisition steps.

[0042] Further, the steps for identifying the types of ship operations may also include the following steps:

[0043] Determine the voyage information data of the ship by using the AIS data and navigation parameters of the bulk cargo ship, including obtaining the starting port and destination port of each voyage of the ship according to the origin and destination port information in the AIS data, and determining the time range of each voyage through the voyage start and end times in the AIS data, so as to determine the basic frame information of the voyage; use the real-time position, longitude and latitude information, course over ground information and AIS information upload time in the AIS data to connect each position point in sequence according to the time sequence, reconstruct the ship's navigation track, and determine the route of the ship in each voyage according to the reconstructed navigation track and combined with the origin and destination port information;

[0044] Based on the AIS data and voyage information data of bulk cargo ships, combined with the route characteristics of bulk cargo ships, long-term charter ships and spot market ships are identified, specifically including: selecting port information data from the database, including port location, port code, port name, and port group code where the port is located; using the port information data and the extracted AIS data of bulk cargo ships for matching, marking the port group codes where the starting and ending ports of each route are located. If the starting and ending ports of some routes are not within the port group, null is assigned; counting the port group codes of the starting and ending ports in the route information, selecting the port group codes of the starting and ending ports that appear the most frequently, calculating the ratio of the fixed routes of a certain ship traveling back and forth between the same starting and ending port groups to all the routes of the ship during the statistical period, and identifying whether the ship is a long-term charter ship or a spot market ship according to the value range of the calculated ratio. The specific steps are as follows:

[0045] (1) Before identifying long-term charter ships and spot market ships, set the time span for identification to 3 months;

[0046] (2) According to the selected time interval, select the AIS data of Chinese coastal dry bulk cargo ships in this time period from the database. The extracted field information includes: mmsi, AIS information upload time, previous AIS information upload time, ship's course over ground information, longitude and latitude information, starting and ending port information, voyage start and end times, etc.;

[0047] (3) Select Chinese coastal port information from the database. The extracted field information includes: port location information, port code, port name, and port group code where the port is located, etc.;

[0048] (4) Use the AIS information upload time, longitude and latitude information, course over ground information, starting and ending port information, and voyage end time in the ship's AIS data to identify the routes of each ship and clarify the destinations of each ship;

[0049] (5) After dividing the routes, use the port information data and the extracted ship AIS data for matching, and mark the port group codes where the starting and ending ports of each route are located. If the starting and ending ports of some routes are not within the Chinese coastal port group, null is assigned;

[0050] (6) Count the port group codes of the starting and ending ports in the route information;

[0051] (7) Select the port group codes of the starting and ending ports that appear the most frequently;

[0052] (8) Preferably, the ratio rate of the fixed routes of Chinese coastal dry bulk cargo ships to all the routes within the preset time can be calculated by the following formula:

[0053]

[0054] Among them, tra jall represents the total number of voyages of a certain ship within the statistical time period, and SE_group_code max represents the number of times (one-way) that the ship travels back and forth between the same starting and ending port groups within the statistical period.

[0055] (9) Preferably, the long-term charter ships and market ships can be identified through the following relational expressions:

[0056]

[0057] Among them, rate is the ratio of the fixed routes of Chinese coastal dry bulk cargo ships to all routes within the preset time, contract_ship is the long-term charter ship, and market_ship is the market ship;

[0058] When rate is greater than or equal to 0.5, it means that more than half of the routes of this ship are fixed routes within the selected time, that is, this ship is a long-term charter ship; when rate is between 0.4 and 0.5, it means that the fixed routes of this ship are relatively few within the selected time, but considering special circumstances, other methods such as visualization need to be used for further judgment (such as drawing the routes and finding relevant captains or chief engineers for judgment); when rate is less than or equal to 0.4, it means that this ship has relatively few fixed routes within the selected time such as three months, that is, this ship is a market ship.

[0059] Exemplarily, in the actual business scenario, shipping companies often do not label long-term charter ships and market ships, which brings certain difficulties to identification and verification. In addition, according to the description of relevant practitioners, the conversion between long-term charter ships and market ships has no regularity, that is, the conversion between long-term charter ships and market ships may occur at any time point, which means that a long-term charter ship can take on the role of a market ship when there is no fixed transportation task. In order to verify the proposed identification logic of long-term charter ships and market ships, in this embodiment, according to the list of long-term charter ships maintained regularly (shown in Table 1), a total of 144 ships, the time periods from April 1, 2022 to July 1, 2022, from August 1, 2022 to October 1, 2022, and from November 1, 2022 to January 1, 2023 are selected as the experimental time periods, and the verification results are shown in Table 2, Table 3, and Table 4 respectively. If the proportion of long-term charter ships within the selected time period exceeds 50%, the identification logic is valid.

[0060] Table 1 Maintained long-term charter ships

[0061] Serial number MMSI (Maritime Mobile Service Identity) 1 4132**000 2 4133**230 … … 143 4773**200 144 4779**600

[0062] Table 2 Verification results in the time period from April 1, 2022 to July 1, 2022

[0063] Serial number MMSI (Maritime Mobile Service Identity) Result Ratio 1 4132**000 Long-term charter vessel 1 2 4132**350 Long-term charter vessel 0.636363636 … … … … 99 4132**660 Long-term charter vessel 0.5 100 4138**000 Market vessel 0.24137931 … … … … 143 4133**470 Market vessel 0.125 144 3529**829 Market vessel 0.117647059

[0064] Table 3 Verification Results during the Period from August 1, 2022 to October 1, 2022

[0065] Serial number MMSI (Maritime Mobile Service Identity) Result Ratio 1 4132**570 Long-term charter vessel 0.681818182 2 4147**000 Long-term charter vessel 0.576923077 … … … … 112 4133**390 Long-term charter vessel 0.5 113 4141**000 Market vessel 0.238095 … … … … 143 4773**200 Market vessel — 144 4779**600 Market vessel —

[0066] Table 4 Verification Results during the Period from November 1, 2022 to January 1, 2023

[0067]

[0068]

[0069] There are 99 ships with a ratio exceeding 0.5 in Table 2, 112 ships with a ratio exceeding 0.5 in Table 3, and 94 ships with a ratio exceeding 0.5 in Table 4. The overall proportions in each table are 68.78%, 77.78%, and 65.28% respectively. This means that these ships are likely to be long-term charter ships. The verification success rate is stable above 65%, indicating that the identification logic of market ships and long-term charter ships proposed in this embodiment is feasible. After identifying long-term charter ships and market ships, the number of candidate ships can be effectively reduced, making the recommendations of the ship-cargo matching model more accurate and efficient.

[0070] III. Ship Matching with Cargo Step. For the scenario of ship matching with cargo, the ship data is used as the set of objects to be matched. The type of goods, time window, loading and unloading ports, volume, and weight attributes are used as restrictive conditions. Combining the daily rent, a target function regarding the transportation charter expenditure is constructed, and based on this, a cargo-ship matching constraint function with multiple constraint conditions including cargo type matching constraint, time window matching constraint, port matching constraint, load and capacity constraint, draft constraint, and market ship constraint is set. Each batch of goods is solved to obtain the set of ships for transporting each batch of goods, and they are sorted in ascending order of the transportation charter cost to construct the ship preference list for each batch of goods, realizing the matching of suitable ships for the goods.

[0071] Furthermore, the target function regarding the transportation charter expenditure is as follows:

[0072] min H*T

[0073] Among them, H refers to the daily rent of the ship (USD / day), T refers to the total number of days of the transportation voyage (days), and min represents taking the minimum value of the expression.

[0074] Furthermore, the total number of days of the transportation voyage includes the number of days for the ship to sail empty from the current location to the loading port of the goods and the number of days from the ship arriving at the loading port of the goods to the ship unloading all the goods.

[0075] The cargo-matching ship is an important shipping operation. It matches ships with batches of goods. In this scenario, the cargo owner usually provides the shipping company with detailed information about the goods, such as the type, quantity, volume, weight, loading and unloading locations and times of the goods. The shipping company then looks for suitable ships based on the information about the goods and arranges for the goods to be transported on the ships. In this process, factors such as the type of ship, cargo capacity, navigation ability, fuel consumption and rental of the ship are all very important considerations. The cargo owner needs to select a suitable ship according to the characteristics of the goods to ensure that the goods can reach the destination within the specified time. At the same time, the cargo owner also needs to consider factors such as the transportation cost of chartering the ship to ensure the economic efficiency of the operation. For bulk commodities, when transporting the goods, the conditions of loading and unloading in full must be met. Therefore, during the transportation process, there will be no situations such as splitting the goods for loading or unloading midway. So, in the embodiments of the present invention, a batch of goods will only be loaded on one ship.

[0076] Preferably, in the cargo-matching ship step, the cargo-matching ship constraint function with multiple constraint conditions set based on the constructed objective function for transportation charter expenses includes a first constraint condition, a second constraint condition, a third constraint condition, a fourth constraint condition, a fifth constraint condition and a sixth constraint condition:

[0077]

[0078] Among them:

[0079] c j represents the j-th batch of goods,

[0080] Type i represents the transportation type of the i-th ship,

[0081] Type j represents the type of the j-th batch of goods,

[0082] contract_ship refers to a long-term charter ship that does not participate in the ship allocation business in the Chinese coastal dry bulk market,

[0083] Weight i represents the total weight (tons) of goods j,

[0084] Weight i represents the maximum deadweight tonnage (tons) of ship i,

[0085] Volumetric j represents the volume (m 3 ) of goods j,

[0086] Space i represents the maximum hold capacity (m 3 ) of ship i,

[0087] draught ij represents the draught (m) of ship i after loading cargo j.

[0088] mindraught lj represents the minimum draught (m) allowed at the destination port of cargo j.

[0089] [ST il ,ET il is the operation time of ship i in port l, including the time required for ship i to perform loading and unloading operations in port l, etc.

[0090] [ST jl ,ET jl is the loading and unloading time of cargo j in port l.

[0091] The first constraint condition described in formula (1) is the cargo type matching constraint for ships whose loaded cargo types include the types of target cargo. The second constraint condition described in formula (2) is the market ship constraint for excluding long-term charter ships. The third constraint conditions described in formulas (3) and (4) are the deadweight and hold capacity constraints that the deadweight of the ship is greater than or equal to the weight of the cargo and the maximum hold capacity of the ship is greater than or equal to the volume of the cargo. The fourth constraint condition described in formula (5) is the draught constraint that the draught depth of the ship meets the requirements of the cargo loading and unloading port. Other constraint conditions may also be included. For example, the fifth constraint condition is the port matching constraint that the unloading port of the cargo is consistent with the shipping route of the ship. The sixth constraint condition described in formula (6) is the time window matching constraint that the loading and unloading time window of the cargo matches the available time window of the ship.

[0092] Furthermore, in the third constraint condition, the deadweight of the ship is greater than or equal to the weight of the cargo and the deadweight of the ship is less than or equal to 1.5 times the weight of the cargo. It can also be understood that formula (3) represents the deadweight constraint for the balance between the deadweight capacity of the ship and the weight of the cargo.

[0093] Furthermore, the draught depth of the ship after loading the cargo can be calculated through the weight of the cargo, the standard seawater density, and the waterplane area (obtained by querying the ship's hydrostatic parameter table). The draught of the ship after loading the cargo needs to be not less than the minimum draught required at the cargo unloading port.

[0094] Furthermore, the operation time of the ship at the loading port and the operation time at the unloading port, the loading operation time of the cargo at the loading port and the unloading operation time of the cargo at the unloading port can be obtained from the business side. The operation time of the ship at the loading port needs to be not greater than the loading operation time of the cargo at the loading port, and the operation time of the ship at the unloading port needs to be not greater than the unloading operation time of the cargo at the unloading port.

[0095] In the business scenario of matching cargo to ship, the embodiment of the present invention can adopt an inexact match between the cargo loading port and the ship's departure port, that is, under the condition that the time window of the ship's departure port plus the time of the ship sailing from the current position to the loading port is not greater than the loading port time window, the departure port of the ship and the cargo loading port can be allowed to be inconsistent. When the cargo loading port and the ship's departure port are inconsistent, the ship needs to sail to the cargo loading port, which results in the ship needing to sail empty for a certain period of time. The number of days the ship sails empty is the ratio of the mileage from the current position of the ship to the cargo loading port to the ship's design speed.

[0096] Furthermore, the empty mileage of the ship can be calculated using the haversine formula.

[0097] 4. The step of matching ships with cargoes is to use cargo data as a candidate object set for the scenario of matching ships with cargoes. The transport type, real-time location, and transport capacity attributes including deadweight tonnage and maximum hold capacity of the ship are used as restriction conditions. Combined with the fuel consumption, an objective function on the fuel consumption expenditure of ship transportation is constructed. Based on this, a ship-cargo matching constraint function is set with multiple constraint conditions including ship type matching constraint, time window matching constraint, empty sailing mileage constraint, transport capacity matching constraint, and draft constraint. The function is solved for each ship to obtain the set of cargoes transported by the ship, and the cargoes are sorted from small to large according to the transport fuel consumption expenditure. A cargo preference list for each ship is constructed to match the ship with suitable cargoes.

[0098] Furthermore, the objective function of ship transportation fuel consumption expenditure is as follows:

[0099] min Fuel_consumption*OP

[0100] Among them, Fuel_comsumption represents the total fuel consumption of the entire transport voyage, OP represents the unit price of crude oil, and min represents taking the minimum value of the expression.

[0101] Furthermore, the total fuel consumption of the transport voyage includes the fuel consumption of the ship sailing empty from the current position to the cargo loading port and the fuel consumption of the ship sailing from the cargo loading port to the cargo unloading port.

[0102] In the business scenario of matching a ship with goods in an embodiment of the present invention, it refers to a business process in which a ship with specific transportation capacity selects suitable goods from existing consigned goods and then the ship performs the transportation task. In the business process of a ship charterer, it is also possible to obtain new transportation capacity due to the occurrence of a ship charter business. At this time, it is necessary to select existing suitable consigned goods to form the transportation task of the ship, so as to avoid idle ship transportation capacity and loss of economic benefits. At this time, the problem faced is matching a ship with goods. In this process, factors such as the type, weight, volume, transportation time, and port conditions of the goods are all very important considerations. The shipowner needs to select suitable goods according to the real-time position and its own transportation capacity of the ship to ensure that the goods can reach the destination within the specified time. At the same time, the shipping company also needs to consider factors such as the transportation cost of the ship to ensure the economic benefits of operation.

[0103] Preferably, the ship-goods matching constraint function set based on the constructed objective function of the ship transportation fuel consumption expenditure includes a seventh constraint condition, an eighth constraint condition, a ninth constraint condition, a tenth constraint condition, and an eleventh constraint condition:

[0104]

[0105] Wherein:

[0106] v i represents the i-th ship,

[0107] Type i represents the transportation type of the i-th ship,

[0108] Type j represents the type of the j-th batch of goods,

[0109] ET represents the earliest time of the ship i or the goods j at the port k,

[0110] LT represents the latest time for the ship i to arrive at the port k,

[0111] DIS kl represents the distance between the port k and the port l,

[0112] V ref represents the designed sailing speed of the ship i,

[0113] DIS noload represents the no-load sailing mileage of the ship i,

[0114] σ represents the threshold value of the no-load sailing mileage,

[0115] Weight j represents the total weight (tons) of the goods j,

[0116] Weighti Denotes the maximum deadweight of vessel i (tons).

[0117] Volumetric j Denotes the total volume of cargo j (m 3 ).

[0118] Space i Refers to the maximum load space of vessel i (m 3 ).

[0119] draught ij Denotes the draught of vessel i after loading cargo j (m).

[0120] mindraught lj Denotes the minimum draught allowed at the destination port of cargo j (m).

[0121] The seventh constraint condition described in Equation (7) is the vessel type matching constraint for selecting cargo types belonging to the transportation type of the target vessel from the candidate cargo set. The eighth constraint condition described in Equations (8) and (9) is the time window matching constraint for the loading and unloading time window of the cargo to match the available time of the vessel. The ninth constraint condition described in Equation (10) is the unloaded voyage mileage constraint for the unloaded voyage mileage of the vessel to be within the unloaded voyage mileage threshold range. The tenth constraint condition described in Equations (11) and (12) is the transportation capacity matching constraint that the deadweight of the vessel is greater than or equal to the weight of the cargo and the maximum load space of the vessel is greater than or equal to the volume of the cargo. The eleventh constraint condition described in Equation (13) is the draught constraint for the draught depth of the vessel to meet the requirements of the cargo loading and unloading port.

[0122] Furthermore, the unloaded voyage mileage of the vessel from the current position to the cargo loading port must be less than the unloaded voyage mileage threshold.

[0123] Furthermore, the draught depth of the vessel after loading the cargo can be calculated through the cargo weight, standard seawater density, and waterplane area (obtained by querying the ship's hydrostatic parameter table).

[0124] Exemplarily, in this embodiment, six vessels are randomly selected from the China coastal dry bulk cargo transportation vessel library (the vessel information is shown in Table 5), and five cargos are randomly selected from the cargo library (the cargo information is shown in Table 6) to verify the accuracy of the method of matching cargo to vessel and vessel to cargo proposed by the present invention.

[0125] Table 5 Basic attributes of the vessels

[0126]

[0127] Table 6 Basic properties of goods

[0128]

[0129]

[0130] The loading ports and unloading ports of the cargo are shown in Table 6. For 6 ships and 5 batches of cargo, the cargo preference list and ship preference list of the shipowner and cargo owner are obtained according to the cargo-to-ship matching method and ship-to-cargo matching method of the present invention, and are sorted according to the fuel expenditure cost and charter expenditure cost, respectively. In the business scenario of matching cargo to ships and matching ships to cargo, it is necessary to query the AIS data of each ship in the corresponding period according to the release time of each batch of cargo, and determine the position of the ship according to the AIS data. The entire voyage includes: sailing empty from the current position of the ship to the cargo loading port, and sailing from the cargo loading port to the cargo unloading port. For the empty section, the empty mileage of the ship can be calculated using the semi-versus-vector formula. Figure 2 As shown in Table 7, the first route is from cargo loading port CNQHD to cargo unloading port CNYZH, the second route is from cargo loading port CNJIT to cargo unloading port CNZJG, the third route is from cargo loading port CNQUA to cargo unloading port CNDAL, the fourth route is from cargo loading port CNHUA to cargo unloading port CNZJG, and the fifth route is from cargo loading port CNFAN to cargo unloading port CNGUA. The mileage from cargo loading port to cargo unloading port (as shown in Table 7) and the empty mileage combined with the ship's fuel consumption per kilometer can calculate the fuel consumption. Finally, the result of matching cargo to ship is obtained, as shown in Table 8, the ship preference list of cargo 1 is ship 5, ship 2, and ship 4, and the ship preference list of cargo 2 to cargo 5 can be obtained by the same logic; the result of matching ship to cargo is shown in Table 9, the cargo preference list of ship 1 is cargo 5 and cargo 3, and the cargo preference list of ship 2 to ship 6 can be obtained by the same logic.

[0131] Table 7 Distances between ports

[0132] Serial number Loading port Unloading port Distance (km) 1 CNQHD CNYZH 755.81 2 CNHUA CNZJG 698.36 3 CNJIT CNZJG 664.67 4 CNQUA CNDAL 954.71 5 CNFAN CNGUA 413.96

[0133] Table 8 Results of cargo-to-ship ratio

[0134]

[0135]

[0136] Table 9 Results by vessel cargo

[0137] Vessel 1 Vessel 2 Vessel 3 Vessel 4 Vessel 5 Vessel 6 Cargo 1 - 2 - 3 2 2 Cargo 2 - 3 - - 3 - Cargo 3 1 1 - 2 4 1 Cargo 4 - 4 - 4 1 - Cargo 5 2 5 1 1 5 -

[0138] One or more embodiments of the present invention further include a step of bilateral matching of ships and goods. For the scenario of bilateral matching of ships and goods, according to the constructed ship preference list and goods preference list, the Gale-Shapley algorithm is used to select the optimal transportation ship for the goods and the optimal transported goods for the ship, so as to achieve bilateral matching of ships and goods.

[0139] Through feature attribute selection and data acquisition, ship operation type identification, and in the scenario of matching goods with ships, a ship-matching goods constraint function with multiple constraints set based on goods attributes and ship data is solved and the goods preference list for each ship is constructed. In the scenario of matching ships with goods, a ship-matching goods constraint function with multiple constraints set based on goods data and ship attributes is solved and the goods preference list for each ship is constructed. The present invention can efficiently find the optimal ship transportation capacity with the best efficiency and benefit for the shipper and the optimal loaded goods with the best revenue for the shipowner based on multi-objective constraints of ships and goods, achieving a win-win situation for both the shipper and the shipowner.

[0140] Based on the same inventive concept, one or more embodiments of this specification further provide a ship and goods matching system for the bulk cargo market based on multi-objective constraints. Since the principle of the problem solved by the ship and goods matching system for the bulk cargo market based on multi-objective constraints is similar to that of the aforementioned ship and goods matching method for the bulk cargo market based on multi-objective constraints, the implementation of the ship and goods matching system for the bulk cargo market based on multi-objective constraints can refer to the implementation of the aforementioned ship and goods matching method for the bulk cargo market based on multi-objective constraints, and the repeated parts will not be elaborated.

[0141] Figure 3 It is a structural block diagram of a ship and goods matching system for the bulk cargo market based on multi-objective constraints provided by one or more embodiments of this specification. As Figure 3 shown, the ship and goods matching system for the bulk cargo market based on multi-objective constraints includes a feature attribute selection and data acquisition module 101, a ship operation type identification module 102, a goods-to-ship module 103, a ship-to-goods module 104, and a bilateral ship and goods matching module 105. Among them,

[0142] The feature attribute selection and data acquisition module 101 is used to screen out the feature attributes of bulk cargo ships and goods with specific relevance in different business scenarios in combination with the feature attributes of bulk cargo ships and goods and the actual business scenarios, and acquire the ship data and cargo data of the feature attributes. Specifically, it obtains multiple bulk cargo ship data from the ship library, including navigation parameters and AIS data. The navigation parameters include the transport type, deadweight tonnage, designed speed, daily rent, fuel consumption, maximum hold capacity, and draft of the ship. The AIS data includes the real-time position, MMSI, AIS information upload time, longitude and latitude information, course over ground information, origin and destination port information, and the start and end times of the voyage of the ship. It obtains multiple batches of cargo data from the cargo library, including the type, weight, volume, loading and unloading ports of the cargo, and the time window including the earliest loading time and the latest unloading time.

[0143] The ship operation type identification module 102 is used to determine the voyage information data of the ship by using the AIS data and navigation parameters of the bulk cargo ship, and identify the long-term charter ship and spot market ship in combination with the route characteristics of the bulk cargo ship based on the AIS data and voyage information data of the bulk cargo ship.

[0144] For the scenario of chartering a ship for a specific cargo, the module 103 takes the ship data as the set of objects to be matched, uses the type, time window, loading and unloading ports, volume, and weight attributes of the cargo as restrictive conditions, combines the daily rent to construct an objective function for the expenditure on chartering a ship for transportation, and based on this, sets a chartering constraint function for a ship for a specific cargo with multiple constraint conditions including cargo type matching constraint, time window matching constraint, port matching constraint, deadweight and hold capacity constraint, draft constraint, and spot market ship constraint. Solve for each batch of cargo to obtain the set of ships for transporting each batch of cargo, sort them in ascending order of the chartering cost for transportation, construct a ship preference list for each batch of cargo, and realize matching suitable ships for the cargo.

[0145] For the scenario of chartering a cargo for a specific ship, the module 104 takes the cargo data as the set of candidate objects, uses the transport type, real-time position, and transport capacity attributes including deadweight tonnage and maximum hold capacity of the ship as restrictive conditions, combines the fuel consumption to construct an objective function for the expenditure on fuel consumption for ship transportation, and based on this, sets a chartering constraint function for a cargo for a specific ship with multiple constraint conditions including ship type matching constraint, time window matching constraint, empty voyage mileage constraint, transport capacity matching constraint, and draft constraint. Solve for each ship to obtain the set of cargoes transported by the ship, sort them in ascending order of the expenditure on fuel consumption for transportation, construct a cargo preference list for each ship, and realize matching suitable cargoes for the ship.

[0146] The cargo-ship bilateral matching module 105 is used for the cargo-ship bilateral matching scenario. According to the constructed ship preference list and cargo preference list, the Gale-Shapley algorithm is adopted to optimize the matching order of ships and cargoes, and the cargo-ship bilateral matching is realized according to the optimized matching result.

[0147] For the ship-cargo matching method and system based on multi-objective constraints proposed by the present invention, in the business scenario of matching ships with cargoes, the module for matching ships with cargoes takes various attributes of the cargo as restrictive conditions, takes the ship data as the set of objects to be matched, combines the daily rent to construct an objective function for the transportation charter expenditure, and sets a ship-cargo constraint function with multiple constraints based on this, and finally constructs a ship preference list for each batch of cargoes; in the business scenario of matching cargoes with ships, the transportation type, real-time position, and transportation capacity attributes including deadweight tonnage and maximum hold capacity of the ship are taken as restrictive conditions, combines the fuel consumption to construct an objective function for the ship transportation fuel consumption expenditure, and sets a cargo-ship constraint function with multiple constraints based on this, and finally constructs a cargo preference list for each ship. Compared with the traditional method, the present invention fully considers the multi-attribute potential connection between ships and cargoes, has advantages in terms of long matching time, transaction success rate, and economic benefits, can efficiently find the most efficient and beneficial ship capacity for the shipper, find the most profitable loaded cargo for the shipowner, and achieve a win-win situation for both the shipper and the shipowner. The scenarios of matching ships with cargoes and matching cargoes with ships automatically complete the solution of the objective function with multiple constraints in the processor, reduce the analysis cost of matching ships with cargoes and matching cargoes with ships, improve the matching efficiency, and make the ship-cargo matching in the bulk cargo market more efficient and economical.

[0148] It should be noted that the above specific implementation manners can enable those skilled in the art to understand the present invention more comprehensively, but do not limit the present invention in any way. Therefore, although this specification has described the present invention in detail with reference to the drawings and embodiments, those skilled in the art should understand that the present invention can still be modified or equivalently replaced. In short, all technical solutions and their improvements that do not depart from the spirit and scope of the present invention should be covered by the protection scope of the patent of the present invention.

Claims

1. A ship-cargo matching method for the bulk cargo market based on multi-objective constraints, characterized in that, It includes the following steps: Feature attribute selection and data acquisition step: Combining the feature attributes of bulk carrier ships and goods and the actual business scenarios, screening out the feature attributes of bulk carrier ships and goods with specific relevance under different business scenarios, and acquiring the ship data and cargo data of the feature attributes. Specifically: obtaining multiple bulk carrier ship data from the ship database, including navigation parameters and AIS data. The navigation parameters include the transportation type, deadweight tonnage, designed speed, daily rent, fuel consumption, maximum hold capacity, and draft of the ship; the AIS data includes the real-time position, MMSI, AIS information upload time, longitude and latitude information, course over ground information, origin and destination port information, and the time of voyage start and end of the ship; obtaining multiple batches of cargo data from the cargo database, including the type, weight, volume, loading and unloading ports of the goods, and the time window including the earliest loading time and the latest unloading time; Ship operation type identification step: Using the AIS data and navigation parameters of bulk carrier ships to determine the voyage information data of the ships, and based on the AIS data and voyage information data of bulk carrier ships, combining the route characteristics of bulk carrier ships, identifying long-term charter ships and spot market ships; Cargo-ship matching step: For the cargo-ship matching scenario, taking the ship data as the set of objects to be matched, using the type, time window, loading and unloading ports, volume, and weight attributes of the goods as restrictive conditions, combining the daily rent to construct an objective function for transportation charter expenditure, and based on this, setting a cargo-ship matching constraint function with multiple constraint conditions including cargo type matching constraint, time window matching constraint, port matching constraint, deadweight and hold capacity constraint, draft constraint, and spot market ship constraint, solving for each batch of goods to obtain the set of ships for transporting each batch of goods, sorting them in ascending order of transportation charter cost, and constructing a ship preference list for each batch of goods to achieve matching suitable ships for the goods; Ship-cargo matching step: For the ship-cargo matching scenario, taking the cargo data as the set of candidate objects, using the transportation type, real-time position, and transportation capacity attributes including deadweight tonnage and maximum hold capacity of the ship as restrictive conditions, combining the fuel consumption to construct an objective function for ship transportation fuel consumption expenditure, and based on this, setting a ship-cargo matching constraint function with multiple constraint conditions including ship type matching constraint, time window matching constraint, empty voyage mileage constraint, transportation capacity matching constraint, and draft constraint, solving for each ship to obtain the set of goods transported by the ship, sorting them in ascending order of transportation fuel consumption expenditure, and constructing a cargo preference list for each ship to achieve matching suitable goods for the ship.

2. The method according to claim 1, wherein The method further includes a ship-cargo bilateral matching step: For the ship-cargo bilateral matching scenario, according to the constructed ship preference list and cargo preference list, using the Gale-Shapley algorithm to optimize the matching order of ships and goods, and based on the optimized matching result, achieving ship-cargo bilateral matching.

3. The method according to claim 1 or 2, characterized in that, In the step of identifying the ship operation type, the voyage information data of the ship is determined by using the AIS data and navigation parameters of the bulk cargo ship, including obtaining the starting port and destination port of each voyage of the ship according to the starting and ending port information in the AIS data, and determining the time range of each voyage through the voyage start and voyage end times in the AIS data, so as to determine the basic frame information of the voyage; using the real-time position, longitude and latitude information, course over ground information and AIS information upload time in the AIS data, connecting each position point in sequence according to the time sequence, reconstructing the navigation track of the ship, and determining the shipping route of the ship in each voyage according to the reconstructed navigation track and the starting and ending port information.

4. The method according to claim 3, wherein In the step of identifying the ship operation type, based on the AIS data and voyage information data of the bulk cargo ship, combined with the route characteristics of the bulk cargo ship, long-term charter ships and spot market ships are identified, specifically including: selecting port information data from the database, including port location, port code, port name and port group code where the port is located; matching the port information data with the extracted AIS data of the bulk cargo ship, marking the port group codes where the starting and ending ports of each route are located, and if the starting and ending ports of some routes are not within the port group, assigning null; counting the port group codes of the starting and ending ports in the route information, selecting the port group codes of the starting and ending ports with the most occurrences, calculating the ratio of the fixed routes of a ship traveling back and forth between the same starting and ending port groups within the statistical period to all the routes of the ship within the statistical period, and identifying whether the ship is a long-term charter ship or a spot market ship according to the value range of the calculated ratio.

5. The method according to claim 1 or 2, characterized in that, In the step of matching ships with goods, the ship matching constraints of multiple constraints set based on the constructed objective function for transportation charter expenses include a first constraint, a second constraint, a third constraint, a fourth constraint, a fifth constraint and a sixth constraint. The first constraint is the cargo type matching constraint for ships whose loaded cargo type includes the types of target goods. The second constraint is the spot market ship constraint excluding long-term charter ships. The third constraint is the deadweight and hold capacity constraint that the deadweight of the ship is greater than or equal to the weight of the goods and the maximum hold capacity of the ship is greater than or equal to the volume of the goods. The fourth constraint is the draft constraint that the draft depth of the ship meets the requirements of the cargo loading and unloading ports. The fifth constraint is the port matching constraint that the unloading port of the goods is consistent with the shipping route of the ship. The sixth constraint is the time window matching constraint that the loading and unloading time window of the goods matches the available time window of the ship.

6. The method according to claim 1 or 2, characterized in that, In the step of matching ships with cargo, the ship matching constraint function set based on the constructed objective function of the fuel consumption expenditure of ship transportation includes the seventh constraint, the eighth constraint, the ninth constraint, the tenth constraint and the eleventh constraint. The seventh constraint is a ship type matching constraint that selects cargo types belonging to the transportation type of the target ship from the candidate cargo set, the eighth constraint is a time window matching constraint that the cargo loading and unloading time window matches the available time of the ship, the ninth constraint is a ship's empty sailing mileage constraint that the empty sailing mileage is within the empty sailing mileage threshold range, the tenth constraint is a transport capacity matching constraint that the deadweight tonnage of the ship is greater than or equal to the weight of the cargo and the maximum cabin capacity of the ship is greater than or equal to the volume of the cargo, and the eleventh constraint is a draft constraint that the draft of the ship meets the requirements of the cargo loading and unloading port.

7. The method according to claim 1 or 2, characterized in that, In the step of matching cargo with ships, the objective function of transport charter expenditure constructed in combination with the daily rent includes the daily rent and the total number of days of the transport voyage, and the total number of days of the transport voyage includes the number of days for the ship to sail empty from the current position to the cargo loading port and the number of days from the ship arriving at the cargo loading port to the ship unloading the cargo.

8. The method according to claim 1 or 2, characterized in that, In the step of matching ships with cargo, the objective function of ship transportation fuel consumption expenditure constructed in combination with the fuel consumption includes the total fuel consumption of the transport voyage and the unit price of each ton of crude oil. The total fuel consumption of the transport voyage includes the fuel consumption of the ship sailing empty from the current position to the cargo loading port and the fuel consumption of the ship sailing from the cargo loading port to the unloading port.

9. A ship-cargo matching system for the bulk cargo market based on multi-objective constraints, characterized in that, It includes a feature attribute selection and data acquisition module, a ship operation type identification module, a cargo-to-ship matching module, and a ship-to-cargo matching module connected in sequence; in, The characteristic attribute selection and data acquisition module is used to combine the characteristic attributes of bulk cargo ships and cargoes and actual business scenarios, screen out the characteristic attributes of bulk cargo ships and cargoes with specific relevance in different business scenarios, and obtain the ship data and cargo data of the characteristic attributes, specifically: obtain multiple bulk cargo ship data from the ship library, including navigation parameters and AIS data, the navigation parameters include the ship's transport type, deadweight tonnage, design speed, daily rent, fuel consumption, maximum tank capacity and draft depth; the AIS data includes the ship's real-time position, MMSI, AIS information upload time, longitude and latitude information, ground heading information, starting and ending port information, and voyage start and end time; obtain multiple batches of cargo data from the cargo library, including the type, weight, volume, loading and unloading port of the cargo, and the time window including the earliest loading time and the latest unloading time; The ship operation type identification module is used to determine the voyage information data of the ship using the AIS data and navigation parameters of the bulk cargo ship, and identify the long-term contract ship and the market ship based on the AIS data and voyage information data of the bulk cargo ship and the route characteristics of the bulk cargo ship; The cargo-for-ship module is used for the cargo-for-ship scenario. It takes ship data as the set of objects to be matched, uses the type of goods, time window, loading and unloading ports, volume, and weight attributes as restrictive conditions, combines the daily rent to construct an objective function for the transportation charter expenditure of the ship, and based on this, sets a cargo-for-ship constraint function including multiple constraint conditions such as cargo type matching constraint, time window matching constraint, port matching constraint, deadweight and hold capacity constraint, draft constraint, and market ship constraint. It solves for each batch of goods to obtain the set of ships for transporting each batch of goods, sorts them in ascending order according to the transportation charter cost, constructs a ship preference list for each batch of goods, and realizes the matching of suitable ships for the goods. The ship-for-cargo module is used for the ship-for-cargo scenario. It takes cargo data as the set of candidate objects, uses the transportation type of the ship, real-time location, and transportation capacity attributes including deadweight tonnage and maximum hold capacity as restrictive conditions, combines the fuel consumption to construct an objective function for the ship transportation fuel consumption expenditure, and based on this, sets a ship-for-cargo constraint function including multiple constraint conditions such as ship type matching constraint, time window matching constraint, empty voyage mileage constraint, transportation capacity matching constraint, and draft constraint. It solves for each ship to obtain the set of goods transported by the ship, sorts them in ascending order according to the transportation fuel consumption expenditure, constructs a cargo preference list for each ship, and realizes the matching of suitable cargo for the ship.

10. The system according to claim 9, wherein The system further includes a ship-cargo bilateral matching module, which is used for the ship-cargo bilateral matching scenario. According to the constructed ship preference list and cargo preference list, it uses the Gale-Shapley algorithm to optimize the matching order of ships and goods, and realizes the ship-cargo bilateral matching according to the optimized matching result.

Images