Market ship intelligent scheduling method and system based on ship and cargo matching and navigational speed optimization

Abstract

The invention provides a market ship intelligent scheduling method and system based on ship-cargo matching and navigational speed optimization, and the method comprises the steps: randomly extracting a plurality of ships and a plurality of batches of cargos, calculating the ship transportation oil consumption cost and the ship rental cost based on the obtained navigation parameters and economic parameters of each ship, and carrying out the ship transportation oil consumption cost and the ship rental cost under the set related constraint conditions. Target functions based on the ship transportation fuel consumption cost and the ship rental cost are constructed respectively; further establishing a ship-cargo matching model and a cargo-ship matching model respectively to obtain a ship preference list and a cargo preference list, carrying out multiple rounds of bilateral matching by adopting a bilateral ship-cargo matching model constructed based on a Gal-Shapley algorithm, and respectively calculating one-to-one matching results of ships and cargoes in each round; and finally, optimizing the navigational speed by adopting a non-dominated sorting genetic algorithm to obtain the optimal navigational speed of each navigational interval, and realizing optimization of the ship navigational performance according to the optimal navigational speed of each navigational interval.

Description

Technical Field

[0001] The present invention relates to the technical field of transportation logistics and shipping management, and in particular to a market ship intelligent scheduling method and system based on ship-cargo matching and speed optimization. Background Art

[0002] The scheduling of ocean cargo is an important issue in the shipping field, which is of great significance for improving voyage efficiency and reducing costs. However, market ship shipping is one of the main modes of global cargo transportation. Unlike liner shipping, it does not rely on fixed service frequencies, and ships can sail at different speeds on different legs of the route as needed. However, compared with liner shipping, there are more research gaps and complexities in the field of market ship shipping. The global warming problem is caused by excessive emissions of greenhouse gases (GHG), which not only threatens the natural ecological balance, but also endangers human food supply and living environment. It is particularly noteworthy that with the rapid growth of global maritime trade, maritime carbon dioxide (CO2) emissions are expected to increase significantly in the coming decades. According to the International Maritime Organization (IMO), in 2007, CO2 emissions from the maritime transport industry increased by 86% compared with 1990, accounting for 3.3% of the global total emissions. The baseline scenario of the study predicts that this emission will increase by 50% to 250% by 2050.

[0003] The carbon emissions of ships are mainly determined by their fuel consumption, which is affected by the sailing speed. This relationship is complex. On the one hand, the relationship between ship fuel consumption and speed can be described by a cubic function. On the other hand, the sailing speed also determines the sailing time of the ship on a specific route, which affects the service quality of the shipping company and the satisfaction of the shipowner. In order to meet the needs of shippers, improve service quality, reduce operating costs and reduce carbon emissions, shipping companies need to optimize the configuration and speed of ships. Therefore, the optimization of sailing speed in market ship shipping has always been a concern for shipping companies.

[0004] The main problems and shortcomings of the existing technology are: Lack of automation and optimization: At present, the market shipping mode mainly relies on manual intervention, including ship-cargo matching, ship receiving order and ship speed management, and requires relevant business personnel to select cargo or ships based on experience. This traditional matching method has problems such as low efficiency, inaccurate matching, high time consumption, and inability to fully optimize the shipping process. In addition, the order of ship receiving cargo usually needs to be arranged manually. This not only wastes time, but also is prone to errors, resulting in delays and inefficiency in the transportation of goods. Carbon emission problem: Navigation speed has an important impact on the carbon emissions of ships, but the existing methods cannot comprehensively consider the complex relationship between carbon emission reduction and ship-cargo matching and speed optimization, so as to effectively reduce carbon emissions. Limited dimensions: When arranging ship and cargo transportation, traditional methods usually only start from the perspective of a single ship and a single cargo, and lack the ability to comprehensively consider a wider range of dimensions, so that more effective ship and cargo matching, navigation speed optimization and carbon emission reduction cannot be achieved.

[0005] Therefore, the existing ship-cargo matching methods and sailing speed management strategies cannot meet the needs of the market ship shipping field. In order to improve transportation efficiency, reduce carbon emissions, and better meet market demand, an innovative technology is urgently needed to solve these challenges in order to improve the efficiency of market ship shipping, reduce carbon emissions, and better meet market demand. Summary of the invention

[0006] In order to solve the problems of low efficiency, inaccurate matching, large time consumption, inability to fully optimize the shipping process, and inability to effectively reduce carbon emissions in the existing ship-cargo matching methods and navigation speed management strategies, the present invention provides a market ship intelligent scheduling method based on ship-cargo matching and navigation speed optimization. Based on multiple selected ships and multiple batches of cargo, as well as the navigation parameters and economic parameters of the ship, the objective functions based on the ship transportation fuel consumption cost and the ship rental cost are respectively constructed by setting relevant constraints, and the Gale-Shapley algorithm is used to perform multiple rounds of bilateral matching of ships and cargoes, and the non-dominated sorting genetic algorithm is used to iteratively select the optimal speed of each navigation section, so as to achieve the lowest fuel consumption of the voyage (correspondingly, the lowest carbon emission of the ship), the lowest charter cost and the lowest total voyage expenditure. The present invention also relates to a market ship intelligent scheduling system based on ship-cargo matching and navigation speed optimization.

[0007] The technical solution of the present invention is as follows:

[0008] A market ship intelligent scheduling method based on ship-cargo matching and speed optimization, characterized by comprising the following steps:

[0009] Parameter acquisition step: randomly select multiple ships and multiple batches of cargo from the ship library and cargo library respectively, and obtain the navigation parameters and economic parameters of each ship, the navigation parameters include the total fuel consumption and total days of the ship transporting cargo, cargo weight, cargo hold capacity, ship draft, mileage and real-time ship position information, and the economic parameters include fuel price and daily rental of the ship;

[0010] Objective function establishment steps: construct decision variables based on the state of the ship transporting cargo, calculate the fuel consumption cost of the ship transporting cargo according to the total fuel consumption of the ship transporting cargo and the fuel price, and calculate the ship leasing cost according to the total number of days for the ship to transport cargo and the daily rent of the ship, and construct the objective function of matching ships with cargo according to the fuel consumption cost of the ship transporting cargo and the decision variables under multiple ship-to-cargo constraint conditions established with the restriction range of cargo weight, the restriction range of cargo hold capacity, the restriction range of time for loading and unloading cargo, the restriction range of ship draft, stable matching of ship and cargo, and the restriction range of empty-load sailing mileage of the ship as constraints; and construct the objective function of matching cargo with ships according to the ship leasing cost and the decision variables under multiple cargo-to-ship constraint conditions established with the restriction range of cargo weight, the restriction range of cargo hold capacity, the restriction range of time for loading and unloading cargo, the restriction range of ship draft, and stable matching of ship and cargo as constraints;

[0011] Ship-cargo matching steps: a ship-cargo matching model is established based on the objective function of matching ship with cargo, multiple ship-cargo matching constraints and real-time ship positions, and a cargo ship matching model is established based on the objective function of matching cargo with ship, multiple cargo-ship matching constraints and real-time ship positions, each batch of cargo is input into the cargo ship matching model, and a ship preference list is obtained that is sorted from large to small according to the ship rental cost; each ship is input into the ship-cargo matching model, and a cargo preference list is obtained that is sorted from large to small according to the fuel consumption expenditure of the ship; then, according to the ship preference list and the cargo preference list, a bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm is used to perform multiple rounds of dynamic bilateral matching, and the one-to-one matching results of each round of ships and cargoes are calculated respectively, and the cargo receiving route of each ship is constructed according to the order of the matching results of each round;

[0012] Initial population generation step: divide the cargo receiving route of each ship into multiple navigation segments according to the cargo receiving order, and randomly generate multiple initial actual speeds for each navigation segment within the actual speed value range, generate the initial population according to all the initial actual speeds in a navigation segment, and treat each actual speed in the initial population as an individual;

[0013] Fitness value calculation and offspring individual generation steps: The fitness value of each individual in the initial population is calculated according to the objective function value, and the genetic operator in the non-dominated sorting genetic algorithm is used to perform neighborhood search evolution operations on multiple individuals in the initial population to generate multiple new offspring individuals, and the fitness value of each new offspring individual is calculated;

[0014] New population generation step: merge the new offspring individuals with the parent individuals to generate a temporary population, and compare the number of individuals in the temporary population with the preset number threshold. If the number of individuals in the temporary population is greater than the preset number threshold, sort the fitness value of each individual in the temporary population by size, and delete the individuals that exceed the number threshold in the sorted temporary population from small to large to obtain a new population;

[0015] Optimal solution calculation steps: Based on the new population, the fitness value calculation and offspring individual generation steps and the new population generation steps are repeated until the preset number of iterations is reached. The individual with the largest fitness value in the new population when the preset number of iterations is reached is taken as the optimal solution of the objective function, and the actual speed corresponding to the optimal solution is taken as the optimal speed of the navigation interval, thereby obtaining the optimal speed for each navigation interval, and optimizing the navigation performance of the ship according to the optimal speed for each navigation interval.

[0016] Preferably, in the ship-cargo matching step, performing multiple rounds of dynamic bilateral matching according to the ship preference list and the cargo preference list and using a bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm specifically includes:

[0017] S1: Multiple ships and multiple batches of cargo are grouped into ship sets and cargo sets respectively. According to the ship preference list and cargo preference list, the bilateral ship-cargo matching model based on the Gale-Shapley algorithm is used to perform the first round of bilateral matching to obtain a one-to-one matching result between ships and cargoes. The matched cargo is removed from the cargo set, while the ship set remains unchanged. After the first round of bilateral matching, the real-time ship position of each ship after the change is recorded.

[0018] S2: A new ship-cargo matching model is established based on the objective function of matching ship with cargo, multiple cargo-ship matching constraints and the changed real-time ship position, and a new cargo-ship matching model is established based on the objective function of matching cargo with ship, multiple cargo-ship matching constraints and the changed real-time ship position, a new ship preference list of the remaining cargo after elimination is calculated according to the new cargo-ship matching model, and a new cargo preference list of the ship after the position change is calculated according to the new ship-cargo matching model; a new round of bilateral matching is performed based on the new ship preference list and the new cargo preference list, and a new round of one-to-one matching results of ships and cargoes are obtained by using the bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm; the matched cargoes are eliminated from the cargo set, the ship set remains unchanged, and after the new round of bilateral matching, the real-time ship position of each ship after the change is recorded;

[0019] S3: Determine whether there are unmatched goods and whether any of the unmatched goods can form a one-to-one matching result with a certain ship. If so, continue to perform the next round of matching according to the principle of step S2; if not, end the matching; after the matching is completed, the one-to-one matching result of each round of ships and goods is obtained, and the successfully matched goods are removed from the goods set, while the ship set remains unchanged.

[0020] Preferably, in the fitness value calculation and offspring individual generation steps, the genetic operator includes a selection operator, a crossover operator and a mutation operator; using the genetic operator in the non-dominated sorting genetic algorithm to perform a neighborhood search evolution operation on multiple individuals in the initial population to generate multiple new offspring individuals specifically includes:

[0021] S1: The selection operator selects multiple individuals as parent individuals from the initial population according to the fitness value and adopts the roulette wheel selection method;

[0022] S2: Determine whether to perform a crossover operation based on the randomly generated probability value and the preset crossover rate. If yes, the crossover operator randomly selects two individuals from the parent individuals, and then randomly selects two speeds with relative positions from the two individuals, and exchanges the codes of these two time units to generate two offspring individuals; if no, the number of parent individuals remains unchanged, and all individuals in the parent individuals are used as offspring individuals;

[0023] S3: Repeat step S2 until the preset number of iterations is reached to obtain multiple offspring individuals;

[0024] S4: The mutation operator performs mutation operations on each offspring individual to obtain multiple new offspring individuals.

[0025] Preferably, in the objective function establishment step, the ship-to-cargo constraint and the cargo-to-ship constraint both include a first constraint, a second constraint, a third constraint, a fourth constraint, a fifth constraint, a sixth constraint, a seventh constraint and an eighth constraint, the first constraint being constructed based on the maximum cargo weight, cargo attributes and decision variables of the ship, the second constraint being constructed based on the cargo hold capacity, cargo attributes and decision variables of the ship, the third constraint being constructed based on the maximum draft that the ship can bear, cargo attributes, decision variables, the maximum draft allowed at the cargo location and the ship leasing status, and the fourth, fifth, sixth, seventh and eighth constraints being all constructed based on the decision variables.

[0026] Preferably, in the objective function establishment step, the ship-cargo constraint condition further includes a ninth constraint condition, and the ninth constraint condition is constructed according to the empty-load sailing mileage and empty-load sailing distance threshold of the ship.

[0027] A market ship intelligent scheduling system based on ship-cargo matching and speed optimization, characterized by comprising a parameter acquisition module, an objective function establishment module, a ship-cargo matching module, an initial population generation module, a fitness value calculation and offspring individual generation module, a new population generation module and an optimal solution calculation module connected in sequence,

[0028] The parameter acquisition module randomly selects multiple ships and multiple batches of cargo from the ship library and the cargo library, and obtains the navigation parameters and economic parameters of each ship. The navigation parameters include the total fuel consumption and total days for the ship to transport cargo, cargo weight, cargo hold capacity, ship draft, mileage and real-time ship position information. The economic parameters include fuel prices and daily rental of the ship.

[0029] The objective function establishment module constructs decision variables based on the state of the ship transporting cargo, calculates the fuel consumption cost of the ship transporting cargo according to the total fuel consumption of the ship transporting cargo and the fuel price, and calculates the ship leasing cost according to the total number of days for the ship to transport cargo and the daily rent of the ship, and constructs the objective function of matching ships with cargo according to the fuel consumption cost of the ship transport and the decision variables under multiple ship-to-cargo constraint conditions established with the restriction range of cargo weight, the restriction range of cargo hold capacity, the time restriction range of ship loading and unloading cargo, the restriction range of ship draft, stable matching of ship and cargo, and the restriction range of empty-load sailing mileage of the ship as constraints; and constructs the objective function of matching cargo with ships according to the ship leasing cost and the decision variables under multiple cargo-to-ship constraint conditions established with the restriction range of cargo weight, the restriction range of cargo hold capacity, the time restriction range of ship loading and unloading cargo, the restriction range of ship draft, and stable matching of ship and cargo as constraints;

[0030] The ship-cargo matching module establishes a ship-cargo matching model based on the objective function of matching ship with cargo, multiple ship-cargo matching constraints and real-time ship positions, and establishes a cargo ship matching model based on the objective function of matching cargo with ship, multiple cargo-ship matching constraints and real-time ship positions, inputs each batch of cargo into the cargo ship matching model, and obtains a ship preference list sorted from large to small according to the ship rental cost; inputs each ship into the ship-cargo matching model, and obtains a cargo preference list sorted from large to small according to the fuel consumption expenditure of the ship; then, according to the ship preference list and the cargo preference list, a bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm is used to perform multiple rounds of dynamic bilateral matching, and the one-to-one matching results of each round of ships and cargoes are calculated respectively, and the cargo receiving route of each ship is constructed according to the order of the matching results of each round;

[0031] The initial population generation module divides the cargo receiving route of each ship into multiple navigation segments according to the cargo receiving sequence, and randomly generates multiple initial actual speeds for each navigation segment within the actual speed value range, generates an initial population based on all the initial actual speeds in a certain navigation segment, and regards each actual speed in the initial population as an individual;

[0032] The fitness value calculation and offspring individual generation module calculates the fitness value of each individual in the initial population according to the objective function value, and uses the genetic operator in the non-dominated sorting genetic algorithm to perform neighborhood search evolution operations on multiple individuals in the initial population to generate multiple new offspring individuals, and calculates the fitness value of each new offspring individual;

[0033] The new population generation module merges the new offspring individuals with the parent individuals to generate a temporary population, and compares the number of individuals in the temporary population with a preset number threshold. If the number of individuals in the temporary population is greater than the preset number threshold, the fitness value of each individual in the temporary population is sorted according to size, and the individuals exceeding the number threshold in the sorted temporary population are deleted in order from small to large to obtain a new population;

[0034] The optimal solution calculation module is based on the new population and repeats the fitness value calculation and the work of the offspring individual generation module and the new population generation module until the preset number of iterations is reached. The individual with the largest fitness value in the new population when the preset number of iterations is reached is used as the optimal solution of the objective function, and the actual speed corresponding to the optimal solution is used as the optimal speed of the navigation interval, thereby obtaining the optimal speed of each navigation interval, and optimizing the navigation performance of the ship according to the optimal speed of each navigation interval.

[0035] Preferably, in the ship cargo matching module, according to the ship preference list and the cargo preference list, a bilateral ship cargo matching model constructed based on the Gale-Shapley algorithm is used to perform multiple rounds of dynamic bilateral matching, which specifically includes:

[0036] S1: Multiple ships and multiple batches of cargo are grouped into ship sets and cargo sets respectively. According to the ship preference list and cargo preference list, the bilateral ship-cargo matching model based on the Gale-Shapley algorithm is used to perform the first round of bilateral matching to obtain a one-to-one matching result between ships and cargoes. The matched cargo is removed from the cargo set, while the ship set remains unchanged. After the first round of bilateral matching, the real-time position of each ship after the change is recorded.

[0037] S2: A new ship-cargo matching model is established based on the objective function of matching ship with cargo, multiple constraints of matching cargo with ship, and the real-time ship position after the change; a new cargo ship matching model is established based on the objective function of matching cargo with ship, multiple constraints of matching cargo with ship, and the real-time ship position after the change; a new ship preference list of the remaining cargo after the elimination is calculated according to the new cargo ship matching model; and a new cargo preference list of the ship after the position change is calculated according to the new ship-cargo matching model; a new round of bilateral matching is performed based on the new ship preference list and the new cargo preference list, and a new round of one-to-one matching results of ships and cargoes are obtained by using the bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm; the matched cargoes are eliminated from the cargo set, and the ship set remains unchanged; and after the new round of bilateral matching, the real-time position of each ship after the change is recorded;

[0038] S3: Determine whether there are unmatched goods and whether any of the unmatched goods can form a one-to-one matching result with a certain ship. If so, continue to perform the next round of matching according to the principle of step S2; if not, end the matching; after the matching is completed, the one-to-one matching result of each round of ships and goods is obtained, and the successfully matched goods are removed from the goods set, while the ship set remains unchanged.

[0039] Preferably, in the fitness value calculation and offspring individual generation module, the genetic operator includes a selection operator, a crossover operator and a mutation operator; using the genetic operator in the non-dominated sorting genetic algorithm to perform a neighborhood search evolution operation on multiple individuals in the initial population to generate multiple new offspring individuals specifically includes:

[0040] S1: The selection operator selects multiple individuals as parent individuals from the initial population according to the fitness value and adopts the roulette wheel selection method;

[0041] S2: Determine whether to perform a crossover operation based on the randomly generated probability value and the preset crossover rate. If yes, the crossover operator randomly selects two individuals from the parent individuals, and then randomly selects two speeds with relative positions from the two individuals, and exchanges the codes of these two time units to generate two offspring individuals; if no, the number of parent individuals remains unchanged, and all individuals in the parent individuals are used as offspring individuals;

[0042] S3: Repeat step S2 until the preset number of iterations is reached to obtain multiple offspring individuals;

[0043] S4: The mutation operator performs mutation operations on each offspring individual to obtain multiple new offspring individuals.

[0044] Preferably, in the objective function establishment module, the ship-to-cargo constraint and the cargo-to-ship constraint both include a first constraint, a second constraint, a third constraint, a fourth constraint, a fifth constraint, a sixth constraint, a seventh constraint and an eighth constraint, the first constraint being constructed based on the maximum cargo weight, cargo attributes and decision variables of the ship, the second constraint being constructed based on the cargo hold capacity, cargo attributes and decision variables of the ship, the third constraint being constructed based on the maximum draft that the ship can bear, cargo attributes, decision variables, the maximum draft allowed at the location of the cargo and the ship leasing status, and the fourth, fifth, sixth, seventh and eighth constraints being all constructed based on the decision variables.

[0045] Preferably, in the objective function building module, the ship-cargo constraint condition further includes a ninth constraint condition, and the ninth constraint condition is constructed according to the empty-load sailing mileage and empty-load sailing distance threshold of the ship.

[0046] The beneficial effects of the present invention are:

[0047] The present invention provides a market ship intelligent scheduling method based on ship-cargo matching and speed optimization, and also a market ship intelligent scheduling system based on dynamic bilateral ship-cargo matching and speed optimization and an application method thereof. First, multiple ships and multiple batches of cargo are selected, and the navigation parameters and economic parameters of each ship are obtained. Under multiple different constraints, an objective function of matching ships with cargo and an objective function of matching cargo with ships are respectively constructed based on the ship transportation fuel consumption cost and the ship rental cost, and then a ship-cargo matching model and a cargo ship matching model are respectively established to obtain a cargo preference list of each ship and a ship preference list of each batch of cargo. Then, according to the ship preference list and the cargo preference list, a bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm is used to perform multiple rounds of dynamic bilateral matching, and the one-to-one matching results of each round of ships and cargoes are automatically calculated, and the cargo receiving route of each ship is automatically constructed according to the order of the matching results of each round. The dynamic bilateral matching effectively improves the matching accuracy through multiple rounds of matching and optimization, helps to ensure that the cargo is arranged on the most suitable ship, thereby improving the transportation efficiency, and through dynamic and real-time matching of ships and cargoes, it can better cope with emergencies, changes in cargo demand and ship position updates. Finally, each ship's cargo receiving route is divided into multiple navigation segments according to the order of cargo, and the initial actual speed is set for each navigation segment within the actual speed value range. The initial population is generated according to all the initial actual speeds, and the non-dominated sorting genetic algorithm is used to calculate the optimal solution of the objective function. The actual speed corresponding to the optimal solution is used as the optimal speed of the navigation segment, and then the optimal speed of each navigation segment is obtained. The optimization of the ship's navigation performance is achieved according to the optimal speed of each navigation segment, which can effectively reduce the ship's fuel consumption, thereby reducing carbon emissions, and help meet environmental protection regulations and sustainable development goals, and can effectively reduce fuel consumption and reduce transportation costs.

[0048] Therefore, the ship matching and scheduling method composed of dynamic bilateral matching, route cargo receiving sequence arrangement and speed optimization based on genetic algorithm in the present invention has several significant advantages over traditional methods: 1. Automation and intelligent decision-making: modern algorithms and technologies are used to automate the decision-making process, reduce manual intervention, and make ship matching and scheduling more intelligent and efficient. 2. Dynamic matching: traditional methods usually use static matching, while the present invention allows dynamic and real-time matching of ships and cargoes, which means better response to emergencies, changes in cargo demand and ship position updates. 3. Multi-objective optimization: the method takes into account multiple objectives, including the lowest fuel consumption of ships and the lowest ship leasing cost. This can help optimize decisions and make them more comprehensive. 4. Carbon emission reduction: speed optimization based on genetic algorithm can reduce fuel consumption of ships, thereby reducing carbon emissions. This helps to meet environmental regulations and sustainable development goals. 5. Matching accuracy: dynamic bilateral matching improves matching accuracy through multiple rounds of matching and optimization. This helps to ensure that cargo is arranged on the most suitable ship, thereby improving transportation efficiency. 6. Cost savings: by reducing fuel consumption, optimizing route cargo receiving sequence and reducing manual intervention, this method can effectively reduce transportation costs. 7. Diversity and exploration: Genetic algorithms introduce diversity, which helps to explore solutions more widely in the search space. This helps to find better solutions. 8. Real-time response: Due to its dynamic nature, the method can respond to changes in the market and environment in real time, making decision-making more flexible. In summary, the ship matching scheduling method through dynamic bilateral matching, route pick-up sequence arrangement and speed optimization based on genetic algorithms has the advantages of improving efficiency, reducing costs, reducing carbon emissions and realizing intelligent decision-making, making it an important tool for the modern shipping industry.

[0049] The present invention is based on a market ship intelligent scheduling system and an application method thereof based on dynamic bilateral ship-cargo matching and speed optimization, and relates to technologies in the field of transportation logistics and shipping management, and particularly covers dynamic bilateral ship-cargo matching technology, ship-based algorithm, speed optimization technology, and market ship intelligent scheduling technology. The application fields of these technologies include, but are not limited to: shipping management: optimizing the coordinated scheduling of ships and cargoes, improving transportation efficiency, and reducing operating costs; logistics and cargo distribution: ensuring that cargoes are delivered to their destinations quickly, efficiently, and sustainably, and reducing cargo retention time; route planning: providing real-time route information so that ships can choose the most economical and environmentally friendly routes while meeting market demand; carbon emission reduction: reducing carbon emissions through ship speed optimization and intelligent matching, and contributing to environmental sustainability; market shipping: providing market ships with more flexible and efficient scheduling and operation solutions to meet changing market demands. In summary, the technical solutions of the present invention are directly applied to transportation logistics, shipping, and market shipping management to improve transportation and logistics efficiency, reduce operating costs, and reduce adverse effects on the environment.

[0050] The present invention utilizes a variety of information sources, such as AIS data of coastal dry bulk carriers, ship information, port spacing and fuel prices of berths and terminals of coastal ports, and comprehensively uses technologies such as real-time ship location query, multi-objective constraints, bilateral matching algorithms and genetic algorithms to construct an efficient algorithm model. This model combines cargo transportation information and relies on a powerful underlying capacity support platform to automate and optimize ship-cargo matching, reduce the need for manual intervention, and improve the accuracy and efficiency of matching. At the same time, based on real-time ship location information, the present invention adopts a multi-objective optimization method, which not only optimizes the sailing speed to reduce carbon emissions, but also ensures that cargo demand is met. Importantly, the present invention also has the ability to automatically determine the order of receiving cargo for ships, further improving the accuracy and efficiency of matching. This comprehensive technical solution provides cargo carriers and capacity parties with decision-making assistance and prediction for matching coastal bulk commodity transactions. With the support of digital means, the present invention can screen out ships that are suitable for the market and can undertake domestic transportation tasks from a large number of ships to form a candidate set of ships. From this candidate set of ships, the present invention finds the optimal ship capacity to meet the needs of the shipowner, and at the same time finds the cargo that can maximize the profit for the shipowner, thus achieving a win-win situation for the shipowner and the shipowner.

[0051] The present invention also relates to a market ship intelligent scheduling system based on ship-cargo matching and speed optimization, which corresponds to the above-mentioned market ship intelligent scheduling method based on ship-cargo matching and speed optimization, and can be understood as a system for realizing the above-mentioned market ship intelligent scheduling method based on ship-cargo matching and speed optimization, including a parameter acquisition module, an objective function establishment module, a ship-cargo matching module, an initial population generation module, a fitness value calculation and offspring individual generation module, a new population generation module and an optimal solution calculation module connected in sequence, each module works in coordination with each other, based on a plurality of selected ships and a plurality of batches of cargo, as well as the navigation parameters and economic parameters of the ship, combined with the characteristic attributes of the ship and the cargo, the actual business The objective functions based on the fuel consumption cost of ship transportation and the ship leasing cost are constructed by setting a number of relevant constraints, and then the ship-cargo matching model and the cargo ship matching model are constructed respectively. The Gale-Shapley algorithm is used to perform multiple rounds of dynamic bilateral matching of ships and cargoes, so as to obtain the one-to-one matching results of ships and cargoes in each round, and then the cargo receiving sequence (cargo receiving route) of each ship is constructed; for the arranged routes, the speed is set, and the non-dominated sorting genetic algorithm is used to iteratively select the optimal speed for each navigation section, so as to achieve the lowest fuel consumption of the voyage (correspondingly, the lowest carbon emission of the ship), the lowest charter cost and the lowest total voyage expenditure. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 It is a flow chart of the market ship intelligent scheduling method based on ship-cargo matching and speed optimization of the present invention.

[0053] Figure 2 It is a flow chart of the non-dominated sorting genetic algorithm of the present invention. DETAILED DESCRIPTION

[0054] The present invention will be described below in conjunction with the accompanying drawings.

[0055] The present invention relates to an intelligent dispatching method for market ships based on ship-cargo matching and speed optimization, aiming to address the technical and efficiency problems existing in market ship shipping and provide a new solution for sustainable and efficient shipping methods. It includes: combining the characteristic attributes of ships and cargoes, actual business scenarios, and the real-time location of ships, setting an objective function based on the ship transportation fuel consumption cost and the ship rental cost, and setting relevant constraints, using the Gale-Shapley algorithm to perform bilateral matching in each round, so as to obtain a one-to-one matching result of each round of ships and cargoes; combining the matching results of each round, constructing the cargo receiving sequence (cargo receiving route) of each ship; setting the speed for the arranged routes, and using genetic algorithms to iteratively select the optimal speed for each navigation section, so as to achieve the lowest fuel consumption of the voyage (correspondingly, the lowest carbon emissions of the ship), the lowest charter cost, and the lowest total voyage expenditure. The flowchart of the method is as follows Figure 1 As shown, the following steps are included in sequence:

[0056] 1. Parameter acquisition steps: randomly select multiple ships from the dry bulk transport ship library along the coast of China, and randomly select multiple batches of cargo from the cargo library, and obtain the navigation parameters and economic parameters of each ship. The navigation parameters include the total fuel consumption and total days for the ship to transport cargo, cargo weight, cargo hold capacity, ship draft, voyage mileage and real-time ship position information. The economic parameters include crude oil prices and daily rental of ships.

[0057] 2. Objective function establishment steps: construct decision variables based on the state of the ship transporting goods, calculate the fuel consumption cost of the ship transport according to the total fuel consumption of the ship transporting goods and the fuel price, and calculate the ship rental cost according to the total number of days the ship transports goods and the daily rent of the ship. Under multiple ship-to-cargo constraint conditions established with the limit range of cargo weight, the limit range of cargo hold capacity, the limit range of time for loading and unloading cargo, the limit range of ship draft, stable matching of ship and cargo, and the limit range of empty-load sailing mileage of the ship as constraints, construct the objective function of matching ships with cargo according to the fuel consumption cost of the ship transport and the decision variables; and under multiple cargo-to-ship constraint conditions established with the limit range of cargo weight, the limit range of cargo hold capacity, the limit range of time for loading and unloading cargo, the limit range of ship draft, and stable matching of ship and cargo as constraints, construct the objective function of matching cargo with ships according to the ship rental cost and the decision variables;

[0058] Specifically, the goal is to minimize the sum of the fuel consumption cost of ship transportation and the ship rental cost. That is, the objective function is to minimize the sum of the fuel consumption cost of ship transportation and the ship rental cost. The decision variable x is constructed based on the state of ship transportation (i.e., the judgment result of whether ship i transports cargo j). ij , the fuel consumption cost of ship transportation is calculated according to the total fuel consumption of ship transportation and the fuel price, and the ship rental cost is calculated according to the total number of days for ship transportation and the daily rent of the ship; then the total objective function is constructed according to the fuel consumption cost of ship transportation, the ship rental cost and the decision variables, which is expressed as follows:

[0059]

[0060] Among them, Fuel_consumption ij represents the total energy consumption of ship i transporting cargo j; OP represents the fuel price; represents the daily rent of ship i; Time ij It represents the time required for ship i to transport cargo j.

[0061] In order to ensure the rationality and feasibility of the matching, a series of constraints need to be considered. The constraints include:

[0062]

[0063] The above parameters are expressed as follows:

[0064] Fuel_consumption ij : Total energy consumption of ship i transporting cargo j;

[0065] OP: fuel price;

[0066] The daily rental rate of vessel i;

[0067] Time ij : The time required for ship i to transport cargo j;

[0068] N: total number of ships available for cargo transportation;

[0069] M: The total number of cargo batches that need to be transported;

[0070] λ: coefficient of matching the maximum load of the ship with the cargo weight;

[0071] The carrying capacity or payload of vessel i, which represents the maximum weight of cargo that the vessel can transport;

[0072] c j : Goods attributes, such as weight or other relevant characteristics;

[0073] The ship's cargo capacity, which represents the maximum volume of cargo that can be stored on board;

[0074] L ij (loading date): the date on which the cargo j can be transported;

[0075] D ij (Deadline): the latest date by which the cargo j must arrive at the destination;

[0076] draught i : The maximum draft or waterline depth that a vessel can withstand, a limit set based on the degree of immersion of the vessel in water;

[0077] H j : The maximum draft or waterline depth permitted at the location of the cargo, which may vary from port to port or location to location;

[0078] T i : transport type of vessel i;

[0079] C j : The type of goods j;

[0080] R i (Vessel Charter Status): A binary indicator (1 or 0) indicating whether the vessel is on liner charter;

[0081] x ij :Ship i transports cargo j, then x ij =1; otherwise 0;

[0082] y ij : Cargo j is transported by ship i, then y ij =1; otherwise 0;

[0083] m i : The empty-load mileage of ship i;

[0084] σ: acceptable empty-handed sailing distance threshold.

[0085] Among them, formula (1) is used to minimize the total fuel consumption and charter costs of the ship's voyage. In formula (2), ships are assigned to transport only the cargo assigned to them, or choose not to transport any cargo. Formula (3) stipulates that each batch of cargo should be transported by a single ship, or not transported. Formula (4) stipulates that a ship is only authorized to transport cargo assigned to it. In addition, formula (5) stipulates that the maximum deadweight of the ship must be equal to or greater than the weight of the cargo to be transported, and the tonnage of the ship must not exceed a multiple of the cargo weight (λ times). Formula (6) details that the ship's cargo hold capacity must be equal to or greater than the volume of the cargo. Formula (7) requires that the ship must arrive at the cargo location before the specified loading date. Formula (8) requires that the ship transports the cargo to the designated port before the deadline. Formula (9) stipulates the draft restrictions of the ship: when empty, the draft of the ship shall not exceed the maximum draft of the cargo location. After loading, the draft of the ship shall not exceed the maximum draft of the cargo destination port. In addition, formula (10) requires that the transported cargo must conform to the transport type of the ship. Formula (11) prohibits ships from participating in liner chartering. Formula (12) aims to minimize the distance that ships travel without load.

[0086] Furthermore, the above process can be divided into two parts, matching ships with cargoes and matching cargoes with ships, according to the different needs of shipowners and cargo owners, which are used to build the ship preference list of cargoes and the ship's cargo preference list respectively. Both processes use constraints to screen suitable objects and score and sort them through objective functions, thereby building a clear preference list for both parties for the subsequent bilateral matching process.

[0087] First, in the ship and cargo problem, since the cargo transportation cost of the ship is determined, the goal is to minimize the navigation fuel cost (i.e., the ship transportation fuel consumption cost) so that the final net profit is optimal when selecting the appropriate cargo composition transportation task. The ship's fuel consumption expenditure is selected as the cost-based objective function. Therefore, under the constraints of the cargo weight limit range, the cargo hold capacity limit range, the ship loading and unloading time limit range, the ship draft limit range, the stable matching of the ship and the cargo, and the empty mileage limit range of the ship, the ship-cargo objective function is constructed according to the ship transportation fuel consumption cost and the decision variables, which is expressed as follows:

[0088]

[0089] As for the problem of matching cargo with ships, since the quantity of cargo is determined and the total revenue is fixed, the goal of selecting a suitable ship to perform the transportation task is to minimize the navigation cost. The ship rental cost is selected as the objective function. Therefore, under the constraints of cargo weight limit, cargo hold capacity limit, ship loading and unloading time limit, ship draft limit and stable matching of ship and cargo, the objective function of cargo matching is constructed according to the ship rental cost and decision variables, which is expressed as follows:

[0090]

[0091] The constraints of matching ships with cargo and matching cargo with ships both include the first constraint, the second constraint, the third constraint, the fourth constraint, the fifth constraint, the sixth constraint, the seventh constraint and the eighth constraint. Among them, the first constraint is based on the maximum cargo weight that the ship can transport. Goods attributesc j and the decision variable x ij The weight of the cargo to be transported shall not exceed the maximum cargo capacity of the vessel and the transport shall not be carried out by a vessel whose tonnage significantly exceeds the weight of the cargo, as indicated below:

[0092]

[0093] The second constraint is based on the cargo hold capacity of the ship The cargo attributes and decision variables are constructed so that the volume of each batch of cargo cannot exceed the maximum cargo hold capacity of the ship, as shown below:

[0094]

[0095] The third constraint is based on the maximum draught that ship i can bear. i , cargo attributes, decision variables, maximum draft H allowed at the cargo location j and the vessel leasing status R i Construct to ensure that the draft of the ship in ballast state (without cargo) does not exceed the maximum draft limit of the port where the cargo is located, and confirm that the draft after loading the cargo batch does not exceed the maximum draft limit of the destination port, as shown below:

[0096]

[0097] The fourth, fifth, sixth, seventh and eighth constraints are constructed based on the decision variables and are expressed as follows:

[0098]

[0099] Formula (18) requires that the ship must transport the cargo before the scheduled loading date. Formula (19) requires that the ship must deliver the cargo to the designated port before the deadline. Formula (20) stipulates that the cargo type must match the cargo type carried by the ship. Formula (21) states that the ship used for transportation must not be liner chartered. Formula (22) specifies that each batch of cargo can only be transported by one ship.

[0100] In addition, the ship-cargo constraint condition also includes a ninth constraint condition, which is based on the empty-load mileage M of the ship. i and the empty sailing distance threshold σ, focusing on minimizing the empty sailing distance of the ship, which is expressed as follows:

[0101]

[0102] The selection of ships and cargoes is limited by a series of equations (constraints). For ships, the equations ensure that they meet the weight and volume requirements of the cargo, that the draft of the ship meets the restrictions of the cargo location and destination port, and that the ship completes the transportation task within a specified time. These constraints ensure that only ships and cargoes that meet the conditions can enter the next stage of the matching process.

[0103] 3. Ship and cargo matching steps: In the ship and cargo matching problem, both processes screen suitable objects through their respective constraints and score and sort them through their respective objective functions, so as to construct a clear preference list for both parties, that is, by scoring and sorting the qualified ships and cargoes, a preference list for ships and cargoes is established. This list reflects the preferences of each ship and cargo and its priority in the matching process, providing a clear basis for subsequent bilateral matching. A ship-cargo matching model is established based on the objective function of matching ship with cargo, multiple ship-cargo matching constraints and real-time ship positions, and a cargo ship matching model is established based on the objective function of matching cargo with ship, multiple cargo-ship matching constraints and real-time ship positions. Each batch of cargo is input into the cargo ship matching model separately, and a ship preference list sorted from large to small according to the ship leasing cost is obtained; each ship is input into the ship-cargo matching model separately, and a cargo preference list sorted from large to small according to the fuel consumption expenditure of the ship is obtained; then, according to the ship preference list and the cargo preference list, a bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm is used to perform multiple rounds of dynamic bilateral matching, and the one-to-one matching results of ships and cargoes in each round are calculated respectively, and the receiving route of each ship is constructed according to the order of the matching results in each round.

[0104] The bilateral matching process is based on multiple rounds of dynamic bilateral matching based on a clear preference list. By comparing the preferences of ships and cargoes, the most suitable matching solution is finally determined. This matching process takes into account the common needs of ships and cargoes, ensuring the best interests of both parties.

[0105] Specifically, 1) Initialization: for each ship, sort its cargo preference list according to the preferred cargo. Initially, each ship is not matched and the status is "unmatched"; for each batch of cargo, sort its ship preference list according to the preferred ship. Initially, each batch of cargo is not matched and the status is "unmatched";

[0106] 2) Actively make suggestions: Each ship selects the first cargo from its unsuggested cargo preference list and makes suggestions to its preferred cargo; each cargo selects the first ship from its unsuggested ship preference list and makes suggestions to its preferred ship; where making a suggestion means that a certain ship is tentatively transporting cargo No. xxx, and not making a suggestion means that the ship has not yet tentatively transported any cargo;

[0107] 3) Approve or reject suggestions: Each batch of cargo compares the received suggestions, selects the most favorable suggestion, and accepts it. If a batch of cargo accepts the suggestion of a ship, its status is changed to "matched", and the status of the ship is also changed to "matched"; after the ship receives the suggestion for the cargo, it selects the most favorable suggestion for it (for example, the cargo preference list of ship A includes cargo No. 1, 2, and 4; cargo No. 1 is the most favorable cargo for ship A, so ship A will accept the suggestion to transport cargo No. 1) and accept it; if a ship accepts the suggestion for a batch of cargo, its status is changed to "matched", and the status of the cargo is also changed to "matched"; if a batch of cargo is rejected or no suggestion is received, continue to make suggestions to the next preferred ship. Similarly, if a ship is rejected or no suggestion is received, continue to make suggestions to the next preferred cargo;

[0108] 4) Loop matching: repeat steps 2 and 3 until no more suggestions can be made or there are no unmatched ships or cargoes;

[0109] 5) Matching ends: Matching ends when no more proposals can be made or there are no unmatched ships or cargoes; each cargo and each ship is matched with an optimal partner, forming stable one-to-one matching results that meet the preferred preferences of each ship and each cargo while avoiding unstable matching.

[0110] Through the above bilateral matching process, each batch of cargo will be matched with the most suitable ship, and each ship will be matched with the most suitable cargo to meet the needs and preferences of both parties. This process will continue to cycle until all cargoes and ships are matched, ensuring the fairness and stability of the matching. The following is a detailed description of the content of the dynamic bilateral cargo matching process with a specific example:

[0111] S1. Initialization:

[0112] Input: m ships, n batches of cargo, real-time ship positions, multi-objective constraints (i.e., ship-to-cargo constraints and cargo-to-ship constraints). For each batch of cargo and each ship, solve the multi-objective constraints to obtain the cargo preference list and ship preference list respectively.

[0113] S2. First round of matching:

[0114] The m ships and n cargoes are grouped into a ship set and a cargo set respectively. According to the ship preference list and the cargo preference list, the Gale Shapley algorithm is used to perform the first round of bilateral matching to obtain a one-to-one ship-cargo matching result. The matched cargo is removed from the cargo set, and the ship set remains unchanged. After the first round of bilateral matching, the real-time ship position of each ship after the change is recorded.

[0115] S3. Loop matching:

[0116] For the remaining cargo and the recorded real-time ship positions after the change, the new cargo preference list and the new ship preference list are re-solved according to the multi-objective constraints. A new round of bilateral matching is performed using the Gale Shapley algorithm to obtain a new round of one-to-one ship-cargo matching results for ships and cargoes, and the matched cargo is removed from the cargo set, while the ship set remains unchanged. After the new round of bilateral matching, the real-time ship position of each ship after the change is recorded;

[0117] S4. Matching judgment: judge whether there are unmatched goods and whether a certain one of the unmatched goods can form a one-to-one matching result with a certain ship. If so, continue to perform the next round of matching according to the principle of step S2; if not, end the matching;

[0118] S5. Matching ends:

[0119] After the matching is completed, the system obtains a series of one-to-one ship-cargo matching results for each round, and each batch of cargo is assigned to a ship while satisfying multi-objective constraints. The successfully matched cargo is removed from the cargo set, while the ship set remains unchanged. The result of each round of matching takes into account the real-time position of the ship and multi-objective constraints to achieve the best ship-cargo matching.

[0120] This process is repeated until there are no more cargoes to be transported or a one-to-one matching result cannot be formed (i.e. until the matching result is null). Through multiple rounds of matching, the system can dynamically adjust the matching between ships and cargoes to meet demand, minimize costs and take into account real-time ship positions. This dynamic bilateral matching method can effectively optimize ship-cargo matching in a constantly changing environment.

[0121] After obtaining the one-to-one matching results of each round through dynamic bilateral matching, the cargo receiving route of each ship is constructed according to the order of the matching results of each round. For example, there are m ships and n batches of cargo, and a total of 4 rounds of matching are carried out. Ship A received cargo C in the first round, cargo B in the second round, cargo H in the third round, and cargo E in the fourth round. Then the cargo receiving order (cargo receiving route) of ship A is: CBHE; ship B received cargo A in the first round and cargo F in the second round. No suitable cargo was matched in the next two rounds. Then the cargo receiving order (cargo receiving route) of ship B is: AF. And so on.

[0122] After constructing the cargo receiving route for each ship, the ship speed is optimized based on the genetic algorithm. In this optimization process, the speed of each voyage segment of the cargo receiving route needs to be considered to minimize fuel costs and carbon emissions while meeting cargo demand. The following is a detailed optimization process:

[0123] 4. Initial population generation steps: Divide each ship's cargo route into multiple navigation segments according to the order of cargo, and randomly generate multiple initial actual speeds for each navigation segment within the actual speed value range. Generate an initial population based on all the initial actual speeds in a certain navigation segment, and treat each actual speed in the initial population as an individual, and set the number of initial populations.

[0124] 5. Fitness value calculation and offspring individual generation step: The fitness value of each individual in the initial population is calculated according to the objective function value, and the genetic operator in the non-dominated sorting genetic algorithm is used to perform neighborhood search evolution operations on multiple individuals in the initial population to generate multiple new offspring individuals, and the fitness value of each new offspring individual is calculated. Preferably, the genetic operator includes a selection operator, a crossover operator, and a mutation operator.

[0125] Specifically, Figure 2 As shown, first define an objective function min(Fuel_consumption*OP+H*T), whose goal is to minimize the fuel cost and charter cost of each cargo receiving route. This objective function will take into account the speed and cargo demand of each voyage segment, and calculate the fitness value of each individual in the initial population according to the objective function value. The higher the fitness value, the better the speed configuration of its voyage segment. Calculate according to the following formula:

[0126]

[0127] Then, a variety of genetic operators are designed to improve and upgrade the individuals in the initial population. Genetic operators can include selection operators, crossover operators, and mutation operators.

[0128] 1. Selection operator: It is the beginning of the evolution operator. It plays a key role in maintaining the continuous inheritance of good individuals and temporarily delaying the spread of poor individuals during the evolution process. The selection operator selects multiple individuals from the initial population as parent individuals based on the fitness value and the roulette selection method. The specific steps are as follows:

[0129] 1) Calculate the probability of each individual being inherited to the next generation, that is, calculate the proportion of the individual fitness value in the overall total fitness value (the total fitness value is the sum of the fitness values ​​of all individuals), and calculate according to the following formula:

[0130]

[0131] Then, the cumulative probability value of each individual is calculated according to the proportion of the individual fitness value in the overall total fitness value (the total fitness value is the sum of the fitness values ​​of all individuals), and the calculation is performed according to the following formula:

[0132]

[0133] where q l Chromosome v j The cumulative probability of .

[0134] 2) Generate a random number in the interval [0,1] and use it as the discrimination value to determine the selected individual. If the random number falls within the cumulative probability value interval of an individual, the individual is selected;

[0135] 3) Add the selected individuals to the candidate improvement set, and repeat the above steps continuously until the number of candidate improvement sets reaches a certain number, and use the candidate improvement set as the parent individual.

[0136] 2. Crossover operator: Follow the steps below:

[0137] 1) First, a probability value is randomly generated (usually a random floating point number between 0 and 1), and then the probability value is compared with the preset crossover rate; when the generated probability value is less than the crossover rate, the crossover operation is selected; when the generated probability value is greater than the crossover rate, the crossover step is directly skipped, the number of parent individuals is kept unchanged, and all individuals in the parent individuals are used as offspring individuals;

[0138] 2) After determining to perform the crossover operation, randomly select two individuals from the parent individuals;

[0139] 3) Randomly select two time units with opposite positions from two individuals, and exchange the codes of these two time units to generate two offspring individuals; for example, as shown in Table A, the code of the eighth time unit is selected for exchange, the eighth time unit of code 1 changes from "1" to "-1", and the eighth time unit of code 2 changes from "-1" to "1";

[0140] 4) Repeat steps 2 and 3 until the preset number of iterations is reached to obtain multiple offspring individuals.

[0141] Table A

[0142]

[0143] It should be noted that the above crossover rate is a set value between 0 and 1, which indicates the probability of performing a crossover operation. For example, if the set crossover rate is 0.8, then in each crossover operation, there is an 80% probability of performing a crossover, and a 20% probability of not performing a crossover, and if a crossover is not performed, the parent individual is directly copied to the child individual.

[0144] 3. Mutation operator: Perform mutation operation on each offspring individual to obtain multiple mutated new offspring individuals, which can prevent the algorithm from falling into local optimality and increase regional diversity. It is a key factor for the algorithm to conduct global exploration. That is, randomly select two time units from a new offspring individual and exchange the codes of the two time units. As shown in Table B, the code -1 of the fifth time unit is exchanged with the code 2 of the third time unit to obtain the mutated new offspring individual. Then calculate the fitness value of each new offspring individual.

[0145] Table B

[0146]

[0147] The mutation operation is an operation used to introduce a certain degree of randomness and diversity to avoid falling into a local optimal solution and explore new solution space. In the optimization of the speed of the ship's cargo pickup sequence, the mutation operation is used to make small, random changes on the existing cargo pickup route to generate new individuals. This is achieved in the present invention through route exchange. This mutation operation can select two different cargo pickup routes and then exchange part of the speed configuration between them. This can introduce greater diversity because it allows information exchange between different cargo pickup routes.

[0148] 6. New population generation step: merge the new offspring individuals with the parent individuals to generate a temporary population. In order to avoid the slow calculation caused by the increasing number of individuals in the temporary population, the number of individuals in the temporary population is compared with the preset number threshold. If the number of individuals in the temporary population is greater than the preset number threshold, the fitness value of each individual in the temporary population is sorted by size, and the individuals exceeding the number threshold in the sorted temporary population are deleted in order from small to large to obtain a new population. This step can be understood as setting a size-maintaining operator for the region. When the number of regions reaches a certain number, they are arranged from small to large according to the fitness value, and the redundant individuals at the end are deleted.

[0149] 7. Optimal solution generation step: Based on the new population, the fitness value calculation and the generation steps of offspring individuals and the generation steps of the new population are repeated until the preset number of iterations is reached. The individual with the largest fitness value in the new population when the preset number of iterations is reached is taken as the optimal solution of the objective function, and the actual speed corresponding to the optimal solution is taken as the optimal speed of the navigation interval, and then the optimal speed of each navigation interval is obtained, and the ship's navigation performance is optimized according to the optimal speed of each navigation interval. That is, once the termination condition is reached, the optimal speed configuration is selected to achieve the lowest fuel cost and the lowest charter expenditure, while meeting the cargo demand.

[0150] The genetic algorithm process of the present invention will continue to iterate and try different speed configurations to find the best solution for each cargo receiving route. By considering multiple factors such as cargo demand, speed, fuel cost and carbon emissions, the speed optimization of the ship's cargo receiving sequence can be achieved to achieve more economical, environmentally friendly and efficient transportation.

[0151] Example:

[0152] Six ships were randomly selected from the China Coastal Dry Bulk Cargo Ship Database, and five cargoes were randomly selected from the cargo database. Their information is shown in Table 1 (Basic Attributes of Ships) and Table 2 (Basic Attributes of Cargoes). The cargo locations and destination ports of the selected cargoes are shown in the figure, and the port spacing between the location and destination ports of each pair of cargoes is shown in Table 3.

[0153] Table 1

[0154]

[0155] Table 2

[0156]

[0157] For 6 ships and 5 cargoes, according to the ship-cargo matching model and the cargo ship matching model, the cargo preference list and the ship preference list of the shipowner and the cargo owner are obtained, and they are sorted according to the fuel consumption expenditure cost and the ship leasing expenditure cost. When using the cargo ship matching model and the ship-cargo matching model, 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 is calculated from this. The entire voyage is divided into three parts: from the empty voyage to the location of the cargo, the cargo loading time and the cargo arrival at the destination, the shipowner's expenditure on leasing the ship, and the shipowner's fuel consumption expenditure will consist of these three parts. The time required to transport the cargo to the destination is calculated based on the port distance and design speed shown in Table 3 (port spacing between port pairs). For the empty voyage, the voyage mileage will be simplified to the straight-line distance between the ship position and the cargo position when the ship undertakes the transportation task, based on which the empty voyage time and the ship's fuel consumption can be calculated.

[0158] Table 3

[0159]

[0160] Table 4 is from the cargo matching model, which is the result of matching cargo to ship (the preference results are shown in rows, and the smaller the number, the higher the priority). For each cargo, the model will output a ship preference list, sorted from small to large by ship rental cost. Among them, Cargo1, Cargo2, Cargo3, Cargo4, and Cargo5 represent different cargoes, and Vessel1, Vessel2, Vessel3, Vessel4, Vessel5, and Vessel6 represent different ships.

[0161] Table 4

[0162]

[0163] Table 5 is from the ship-cargo matching model and is the result of matching ships with cargo (preference results are shown in columns, the smaller the number, the higher the priority). For each ship, the model outputs a list of cargo preferences, sorted from small to large according to the ship's fuel consumption expenditure.

[0164] Table 5

[0165]

[0166] After obtaining the preference lists of ship owners and cargo owners respectively, the bilateral matching model based on the Gale-Shapley algorithm is used to match ships and cargoes. The results of bilateral matching are shown in Table 6.

[0167] Table 6

[0168]

[0169] As can be seen from Table 6, the bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm yields the following results: The first batch of goods will be transported by Ship 4, the third batch of goods will be transported by Ship 2, the fourth batch of goods will be transported by Ship 5, and the fifth batch of goods will be transported by Ship 1. There is still Goods 2 left without a ship for transportation. The real-time positions of the 6 ships are queried again, and a new round of construction of the goods preference list of the ships and the ship preference list of the goods is carried out. The results are shown in Table 7 and Table 8 below. Among them, Table 7 is the result of matching goods to ships in the second round (the preference results are shown in rows, and the smaller the number, the higher the priority), and Table 8 is the result of matching ships to goods in the second round (the preference results are shown in columns, and the smaller the number, the higher the priority).

[0170] Table 7

[0171]

[0172] Table 8

[0173]

[0174]

[0175] After obtaining the preference lists of ship owners and cargo owners respectively in the second round, the bilateral matching model based on the Gale-Shapley algorithm is used for ship-cargo matching. The results of the second-round bilateral matching are shown in Table 9.

[0176] Table 9

[0177]

[0178] As can be seen from Table 9, the last batch of Goods 2 is picked up by Ship 5. At this point, all the goods have been transported away, and the matching stops. According to the receiving order, the receiving routes of each ship are constructed, as shown in Table 10 (Receiving Route Arrangement).

[0179] Table 10

[0180] Receiving route Vessel1 5 Vessel2 3 Vessel3 - Vessel4 1 Vessel5 4-2 Vessel6 -

[0181] After obtaining the receiving routes of each ship, the genetic algorithm is used to optimize the ship speeds of the ships with transportation tasks, so as to minimize the ship fuel consumption and ship rental expenses of the entire fleet. The results are shown in Table 11 below. The results under the initially set ship speeds are shown in Table 12.

[0182] Table 11

[0183]

[0184]

[0185] Table 12

[0186]

[0187] In summary, by optimizing the speed of ships, routes, and the matching and order of cargo, transportation costs, including fuel consumption and ship chartering expenses, can be significantly reduced, thereby improving transportation efficiency and reducing the total cost of cargo transportation. This optimization method has important application potential in scenarios with multiple ships and multiple cargoes, helping to achieve more efficient cargo transportation and cost savings.

[0188] The present invention also relates to a market ship intelligent scheduling system based on ship-cargo matching and speed optimization, which corresponds to the above-mentioned market ship intelligent scheduling method based on ship-cargo matching and speed optimization, and can be understood as a system for implementing the above-mentioned method. The system includes a parameter acquisition module, an objective function establishment module, a ship-cargo matching module, an initial population generation module, a fitness value calculation and offspring individual generation module, a new population generation module and an optimal solution calculation module connected in sequence. Specifically,

[0189] The parameter acquisition module randomly selects multiple ships and multiple batches of cargo from the ship library and the cargo library, and obtains the navigation parameters and economic parameters of each ship. The navigation parameters include the total fuel consumption and total days for the ship to transport cargo, cargo weight, cargo hold capacity, ship draft, mileage and real-time ship position information. The economic parameters include fuel prices and daily rental of the ship.

[0190] The objective function establishment module constructs decision variables based on the state of the ship transporting cargo, calculates the fuel consumption cost of the ship transporting cargo according to the total fuel consumption of the ship transporting cargo and the fuel price, and calculates the ship leasing cost according to the total number of days for the ship to transport cargo and the daily rent of the ship, and constructs the objective function of matching ships with cargo according to the fuel consumption cost of the ship transport and the decision variables under multiple ship-to-cargo constraint conditions established with the restriction range of cargo weight, the restriction range of cargo hold capacity, the time restriction range of ship loading and unloading cargo, the restriction range of ship draft, stable matching of ship and cargo, and the restriction range of empty-load sailing mileage of the ship as constraints; and constructs the objective function of matching cargo with ships according to the ship leasing cost and the decision variables under multiple cargo-to-ship constraint conditions established with the restriction range of cargo weight, the restriction range of cargo hold capacity, the time restriction range of ship loading and unloading cargo, the restriction range of ship draft, and stable matching of ship and cargo as constraints;

[0191] The ship-cargo matching module establishes a ship-cargo matching model based on the objective function of matching ship with cargo, multiple ship-cargo matching constraints and real-time ship positions, and establishes a cargo ship matching model based on the objective function of matching cargo with ship, multiple cargo-ship matching constraints and real-time ship positions, inputs each batch of cargo into the cargo ship matching model, and obtains a ship preference list sorted from large to small according to the ship rental cost; inputs each ship into the ship-cargo matching model, and obtains a cargo preference list sorted from large to small according to the fuel consumption expenditure of the ship; then, according to the ship preference list and the cargo preference list, a bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm is used to perform multiple rounds of dynamic bilateral matching, and the one-to-one matching results of each round of ships and cargoes are calculated respectively, and the cargo receiving route of each ship is constructed according to the order of the matching results of each round;

[0192] The initial population generation module divides the cargo receiving route of each ship into multiple navigation segments according to the cargo receiving sequence, and randomly generates multiple initial actual speeds for each navigation segment within the actual speed value range, generates an initial population based on all the initial actual speeds in a certain navigation segment, and regards each actual speed in the initial population as an individual;

[0193] The fitness value calculation and offspring individual generation module calculates the fitness value of each individual in the initial population according to the objective function value, and uses the genetic operator in the non-dominated sorting genetic algorithm to perform neighborhood search evolution operations on multiple individuals in the initial population to generate multiple new offspring individuals, and calculates the fitness value of each new offspring individual;

[0194] The new population generation module merges the new offspring individuals with the parent individuals to generate a temporary population, and compares the number of individuals in the temporary population with a preset number threshold. If the number of individuals in the temporary population is greater than the preset number threshold, the fitness value of each individual in the temporary population is sorted according to size, and the individuals exceeding the number threshold in the sorted temporary population are deleted in order from small to large to obtain a new population;

[0195] The optimal solution calculation module is based on the new population and repeats the fitness value calculation and the work of the offspring individual generation module and the new population generation module until the preset number of iterations is reached. The individual with the largest fitness value in the new population when the preset number of iterations is reached is used as the optimal solution of the objective function, and the actual speed corresponding to the optimal solution is used as the optimal speed of the navigation interval, thereby obtaining the optimal speed of each navigation interval, and optimizing the navigation performance of the ship according to the optimal speed of each navigation interval.

[0196] Preferably, in the ship cargo matching module, according to the ship preference list and the cargo preference list, a bilateral ship cargo matching model constructed based on the Gale-Shapley algorithm is used to perform multiple rounds of dynamic bilateral matching, which specifically includes:

[0197] S1: Multiple ships and multiple batches of cargo are grouped into ship sets and cargo sets respectively. According to the ship preference list and cargo preference list, the bilateral ship-cargo matching model based on the Gale-Shapley algorithm is used to perform the first round of bilateral matching to obtain a one-to-one matching result between ships and cargoes. The matched cargo is removed from the cargo set, while the ship set remains unchanged. After the first round of bilateral matching, the real-time position of each ship after the change is recorded.

[0198] S2: A new ship-cargo matching model is established based on the objective function of matching ship with cargo, multiple constraints of matching cargo with ship, and the real-time ship position after the change; a new cargo ship matching model is established based on the objective function of matching cargo with ship, multiple constraints of matching cargo with ship, and the real-time ship position after the change; a new ship preference list of the remaining cargo after the elimination is calculated according to the new cargo ship matching model; and a new cargo preference list of the ship after the position change is calculated according to the new ship-cargo matching model; a new round of bilateral matching is performed based on the new ship preference list and the new cargo preference list, and a new round of one-to-one matching results of ships and cargoes are obtained by using the bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm; the matched cargoes are eliminated from the cargo set, and the ship set remains unchanged; and after the new round of bilateral matching, the real-time position of each ship after the change is recorded;

[0199] S3: Determine whether there are unmatched goods and whether any of the unmatched goods can form a one-to-one matching result with a certain ship. If so, continue to perform the next round of matching according to the principle of step S2; if not, end the matching; after the matching is completed, the one-to-one matching result of each round of ships and goods is obtained, and the successfully matched goods are removed from the goods set, while the ship set remains unchanged.

[0200] Preferably, the parameter acquisition module also acquires port parameters, which include the initial inventory of the loading port and the loading capacity of the loading port at a certain moment.

[0201] Preferably, in the fitness value calculation and offspring individual generation module, the genetic operator includes a selection operator, a crossover operator and a mutation operator; using the genetic operator in the non-dominated sorting genetic algorithm to perform a neighborhood search evolution operation on multiple individuals in the initial population to generate multiple new offspring individuals specifically includes:

[0202] S1: The selection operator selects multiple individuals as parent individuals from the initial population according to the fitness value and adopts the roulette wheel selection method;

[0203] S2: Determine whether to perform a crossover operation based on the randomly generated probability value and the preset crossover rate. If yes, the crossover operator randomly selects two individuals from the parent individuals, and then randomly selects two speeds with relative positions from the two individuals, and exchanges the codes of these two time units to generate two offspring individuals; if no, the number of parent individuals remains unchanged, and all individuals in the parent individuals are used as offspring individuals;

[0204] S3: Repeat step S2 until the preset number of iterations is reached to obtain multiple offspring individuals;

[0205] S4: The mutation operator performs mutation operations on each offspring individual to obtain multiple new offspring individuals.

[0206] Preferably, in the objective function establishment module, the ship-to-cargo constraint and the cargo-to-ship constraint both include a first constraint, a second constraint, a third constraint, a fourth constraint, a fifth constraint, a sixth constraint, a seventh constraint and an eighth constraint, the first constraint being constructed based on the maximum cargo weight, cargo attributes and decision variables of the ship, the second constraint being constructed based on the cargo hold capacity, cargo attributes and decision variables of the ship, the third constraint being constructed based on the maximum draft that the ship can bear, cargo attributes, decision variables, the maximum draft allowed at the location of the cargo and the ship leasing status, and the fourth, fifth, sixth, seventh and eighth constraints being all constructed based on decision variables.

[0207] Preferably, in the objective function establishment module, the ship-cargo constraint condition further includes a ninth constraint condition, and the ninth constraint condition is constructed according to the empty-load sailing mileage and empty-load sailing distance threshold of the ship.

[0208] The present invention provides an objective and scientific market ship intelligent scheduling method and system based on ship-cargo matching and speed optimization. Based on multiple selected ships and multiple batches of cargo, as well as the navigation parameters and economic parameters of the ships, objective functions based on the ship transportation fuel consumption cost and the ship rental cost are respectively constructed by setting relevant constraints, and the Gale-Shapley algorithm is used to perform multiple rounds of bilateral matching of ships and cargoes, so as to obtain a one-to-one matching result of ships and cargoes in each round, and then construct the cargo receiving sequence (cargo receiving route) of each ship; for the arranged routes, the speed is set, and the non-dominated sorting genetic algorithm is used to iteratively select the optimal speed of each navigation section, so as to achieve the lowest fuel consumption of the voyage (correspondingly, the lowest carbon emission of the ship), the lowest charter cost and the lowest total voyage expenditure.

[0209] It should be noted that the above-described specific implementations can enable those skilled in the art to more fully understand the invention, but do not limit the invention in any way. Therefore, although this specification has described the invention in detail with reference to the drawings and embodiments, those skilled in the art should understand that the invention can still be modified or replaced by equivalents. In short, all technical solutions and improvements that do not deviate from the spirit and scope of the invention should be included in the protection scope of the patent for the invention.

Claims

1. A market ship intelligent scheduling method based on ship-cargo matching and speed optimization, characterized in that: The following steps are involved: Parameter acquisition step: randomly select multiple ships and multiple batches of cargo from the ship library and cargo library respectively, and obtain the navigation parameters and economic parameters of each ship, the navigation parameters include the total fuel consumption and total days of the ship transporting cargo, cargo weight, cargo hold capacity, ship draft, mileage and real-time ship position information, and the economic parameters include fuel price and daily rental of the ship; Objective function establishment steps: construct decision variables based on the state of the ship transporting cargo, calculate the fuel consumption cost of the ship transporting cargo according to the total fuel consumption of the ship transporting cargo and the fuel price, and calculate the ship leasing cost according to the total number of days for the ship to transport cargo and the daily rent of the ship, and construct the objective function of matching ships with cargo according to the fuel consumption cost of the ship transporting cargo and the decision variables under multiple ship-to-cargo constraint conditions established with the restriction range of cargo weight, the restriction range of cargo hold capacity, the restriction range of time for loading and unloading cargo, the restriction range of ship draft, stable matching of ship and cargo, and the restriction range of empty-load sailing mileage of the ship as constraints; and construct the objective function of matching cargo with ships according to the ship leasing cost and the decision variables under multiple cargo-to-ship constraint conditions established with the restriction range of cargo weight, the restriction range of cargo hold capacity, the restriction range of time for loading and unloading cargo, the restriction range of ship draft, and stable matching of ship and cargo as constraints; Ship-cargo matching steps: a ship-cargo matching model is established based on the objective function of matching ship with cargo, multiple ship-cargo matching constraints and real-time ship positions, and a cargo ship matching model is established based on the objective function of matching cargo with ship, multiple cargo-ship matching constraints and real-time ship positions, each batch of cargo is input into the cargo ship matching model, and a ship preference list is obtained that is sorted from large to small according to the ship rental cost; each ship is input into the ship-cargo matching model, and a cargo preference list is obtained that is sorted from large to small according to the fuel consumption expenditure of the ship; then, according to the ship preference list and the cargo preference list, a bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm is used to perform multiple rounds of dynamic bilateral matching, and the one-to-one matching results of each round of ships and cargoes are calculated respectively, and the cargo receiving route of each ship is constructed according to the order of the matching results of each round; Initial population generation step: divide the cargo receiving route of each ship into multiple navigation segments according to the cargo receiving order, and randomly generate multiple initial actual speeds for each navigation segment within the actual speed value range, generate the initial population according to all the initial actual speeds in a navigation segment, and treat each actual speed in the initial population as an individual; Fitness value calculation and offspring individual generation steps: The fitness value of each individual in the initial population is calculated according to the objective function value, and the genetic operator in the non-dominated sorting genetic algorithm is used to perform neighborhood search evolution operations on multiple individuals in the initial population to generate multiple new offspring individuals, and the fitness value of each new offspring individual is calculated; New population generation step: merge the new offspring individuals with the parent individuals to generate a temporary population, and compare the number of individuals in the temporary population with the preset number threshold. If the number of individuals in the temporary population is greater than the preset number threshold, sort the fitness value of each individual in the temporary population by size, and delete the individuals that exceed the number threshold in the sorted temporary population from small to large to obtain a new population; Optimal solution calculation steps: Based on the new population, the fitness value calculation and offspring individual generation steps and the new population generation steps are repeated until the preset number of iterations is reached. The individual with the largest fitness value in the new population when the preset number of iterations is reached is taken as the optimal solution of the objective function, and the actual speed corresponding to the optimal solution is taken as the optimal speed of the navigation interval, thereby obtaining the optimal speed for each navigation interval, and optimizing the navigation performance of the ship according to the optimal speed for each navigation interval.

2. The market ship intelligent scheduling method based on ship-cargo matching and speed optimization according to claim 1 is characterized in that: In the ship-cargo matching step, multiple rounds of dynamic bilateral matching are performed according to the ship preference list and the cargo preference list and using a bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm, specifically including: S1: Multiple ships and multiple batches of cargo are grouped into ship sets and cargo sets respectively. According to the ship preference list and cargo preference list, the bilateral ship-cargo matching model based on the Gale-Shapley algorithm is used to perform the first round of bilateral matching to obtain a one-to-one matching result between ships and cargoes. The matched cargo is removed from the cargo set, while the ship set remains unchanged. After the first round of bilateral matching, the real-time ship position of each ship after the change is recorded. S2: A new ship-cargo matching model is established based on the objective function of matching ship with cargo, multiple cargo-ship matching constraints and the changed real-time ship position, and a new cargo-ship matching model is established based on the objective function of matching cargo with ship, multiple cargo-ship matching constraints and the changed real-time ship position, a new ship preference list of the remaining cargo after elimination is calculated according to the new cargo-ship matching model, and a new cargo preference list of the ship after the position change is calculated according to the new ship-cargo matching model; a new round of bilateral matching is performed based on the new ship preference list and the new cargo preference list, and a new round of one-to-one matching results of ships and cargoes are obtained by using the bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm; the matched cargoes are eliminated from the cargo set, the ship set remains unchanged, and after the new round of bilateral matching, the real-time ship position of each ship after the change is recorded; S3: Determine whether there are unmatched goods and whether any of the unmatched goods can form a one-to-one matching result with a certain ship. If so, continue to perform the next round of matching according to the principle of step S2; if not, end the matching; after the matching is completed, the one-to-one matching result of each round of ships and goods is obtained, and the successfully matched goods are removed from the goods set, while the ship set remains unchanged.

3. The market ship intelligent scheduling method based on ship-cargo matching and speed optimization according to claim 1 is characterized in that: In the fitness value calculation and offspring individual generation steps, the genetic operators include a selection operator, a crossover operator and a mutation operator; using the genetic operators in the non-dominated sorting genetic algorithm to perform neighborhood search evolution operations on multiple individuals in the initial population to generate multiple new offspring individuals specifically includes: S1: The selection operator selects multiple individuals as parent individuals from the initial population according to the fitness value and adopts the roulette wheel selection method; S2: Determine whether to perform a crossover operation based on the randomly generated probability value and the preset crossover rate. If yes, the crossover operator randomly selects two individuals from the parent individuals, and then randomly selects two speeds with relative positions from the two individuals, and exchanges the codes of these two time units to generate two offspring individuals; if no, the number of parent individuals remains unchanged, and all individuals in the parent individuals are used as offspring individuals; S3: Repeat step S2 until the preset number of iterations is reached to obtain multiple offspring individuals; S4: The mutation operator performs mutation operations on each offspring individual to obtain multiple new offspring individuals.

4. The market ship intelligent scheduling method based on ship-cargo matching and speed optimization according to claim 1 is characterized in that: In the objective function establishment step, the ship-to-cargo constraint and the cargo-to-ship constraint both include a first constraint, a second constraint, a third constraint, a fourth constraint, a fifth constraint, a sixth constraint, a seventh constraint and an eighth constraint. The first constraint is constructed based on the maximum cargo weight, cargo attributes and decision variables of the ship. The second constraint is constructed based on the cargo hold capacity, cargo attributes and decision variables of the ship. The third constraint is constructed based on the maximum draft that the ship can bear, cargo attributes, decision variables, the maximum draft allowed at the cargo location and the ship leasing status. The fourth, fifth, sixth, seventh and eighth constraints are all constructed based on the decision variables.

5. The market ship intelligent scheduling method based on ship-cargo matching and speed optimization according to claim 4 is characterized in that: In the objective function establishment step, the ship-cargo constraint condition further includes a ninth constraint condition, and the ninth constraint condition is constructed according to the empty-load sailing mileage and empty-load sailing distance threshold of the ship.

6. A market ship intelligent dispatching system based on ship-cargo matching and speed optimization, characterized in that: It includes a parameter acquisition module, an objective function establishment module, a ship cargo matching module, an initial population generation module, a fitness value calculation and offspring individual generation module, a new population generation module and an optimal solution calculation module, which are connected in sequence. The parameter acquisition module randomly selects multiple ships and multiple batches of cargo from the ship library and the cargo library, and obtains the navigation parameters and economic parameters of each ship. The navigation parameters include the total fuel consumption and total days for the ship to transport cargo, cargo weight, cargo hold capacity, ship draft, mileage and real-time ship position information. The economic parameters include fuel prices and daily rental of the ship. The objective function establishment module constructs decision variables based on the state of the ship transporting cargo, calculates the fuel consumption cost of the ship transporting cargo according to the total fuel consumption of the ship transporting cargo and the fuel price, and calculates the ship leasing cost according to the total number of days for the ship to transport cargo and the daily rent of the ship, and constructs the objective function of matching ships with cargo according to the fuel consumption cost of the ship transport and the decision variables under multiple ship-to-cargo constraint conditions established with the restriction range of cargo weight, the restriction range of cargo hold capacity, the time restriction range of ship loading and unloading cargo, the restriction range of ship draft, stable matching of ship and cargo, and the restriction range of empty-load sailing mileage of the ship as constraints; and constructs the objective function of matching cargo with ships according to the ship leasing cost and the decision variables under multiple cargo-to-ship constraint conditions established with the restriction range of cargo weight, the restriction range of cargo hold capacity, the time restriction range of ship loading and unloading cargo, the restriction range of ship draft, and stable matching of ship and cargo as constraints; The ship-cargo matching module establishes a ship-cargo matching model based on the objective function of matching ship with cargo, multiple ship-cargo matching constraints and real-time ship positions, and establishes a cargo ship matching model based on the objective function of matching cargo with ship, multiple cargo-ship matching constraints and real-time ship positions, inputs each batch of cargo into the cargo ship matching model, and obtains a ship preference list sorted from large to small according to the ship rental cost; inputs each ship into the ship-cargo matching model, and obtains a cargo preference list sorted from large to small according to the fuel consumption expenditure of the ship; then, according to the ship preference list and the cargo preference list, a bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm is used to perform multiple rounds of dynamic bilateral matching, and the one-to-one matching results of each round of ships and cargoes are calculated respectively, and the cargo receiving route of each ship is constructed according to the order of the matching results of each round; The initial population generation module divides the cargo receiving route of each ship into multiple navigation segments according to the cargo receiving sequence, and randomly generates multiple initial actual speeds for each navigation segment within the actual speed value range, generates an initial population based on all the initial actual speeds in a certain navigation segment, and regards each actual speed in the initial population as an individual; The fitness value calculation and offspring individual generation module calculates the fitness value of each individual in the initial population according to the objective function value, and uses the genetic operator in the non-dominated sorting genetic algorithm to perform neighborhood search evolution operations on multiple individuals in the initial population to generate multiple new offspring individuals, and calculates the fitness value of each new offspring individual; The new population generation module merges the new offspring individuals with the parent individuals to generate a temporary population, and compares the number of individuals in the temporary population with a preset number threshold. If the number of individuals in the temporary population is greater than the preset number threshold, the fitness value of each individual in the temporary population is sorted according to size, and the individuals exceeding the number threshold in the sorted temporary population are deleted in order from small to large to obtain a new population; The optimal solution calculation module is based on the new population and repeats the fitness value calculation and the work of the offspring individual generation module and the new population generation module until the preset number of iterations is reached. The individual with the largest fitness value in the new population when the preset number of iterations is reached is used as the optimal solution of the objective function, and the actual speed corresponding to the optimal solution is used as the optimal speed of the navigation interval, thereby obtaining the optimal speed of each navigation interval, and optimizing the navigation performance of the ship according to the optimal speed of each navigation interval.

7. The market ship intelligent scheduling system based on ship-cargo matching and speed optimization according to claim 6 is characterized in that: In the ship cargo matching module, according to the ship preference list and the cargo preference list, a bilateral ship cargo matching model based on the Gale-Shapley algorithm is used to perform multiple rounds of dynamic bilateral matching, specifically including: S1: Multiple ships and multiple batches of cargo are grouped into ship sets and cargo sets respectively. According to the ship preference list and cargo preference list, the bilateral ship-cargo matching model based on the Gale-Shapley algorithm is used to perform the first round of bilateral matching to obtain a one-to-one matching result between ships and cargoes. The matched cargo is removed from the cargo set, while the ship set remains unchanged. After the first round of bilateral matching, the real-time position of each ship after the change is recorded. S2: A new ship-cargo matching model is established based on the objective function of matching ship with cargo, multiple constraints of matching cargo with ship, and the real-time ship position after the change; a new cargo ship matching model is established based on the objective function of matching cargo with ship, multiple constraints of matching cargo with ship, and the real-time ship position after the change; a new ship preference list of the remaining cargo after the elimination is calculated according to the new cargo ship matching model; and a new cargo preference list of the ship after the position change is calculated according to the new ship-cargo matching model; a new round of bilateral matching is performed based on the new ship preference list and the new cargo preference list, and a new round of one-to-one matching results of ships and cargoes are obtained by using the bilateral ship-cargo matching model constructed based on the Gale-Shapley algorithm; the matched cargoes are eliminated from the cargo set, and the ship set remains unchanged; and after the new round of bilateral matching, the real-time position of each ship after the change is recorded; S3: Determine whether there are unmatched goods and whether any of the unmatched goods can form a one-to-one matching result with a certain ship. If so, continue to perform the next round of matching according to the principle of step S2; if not, end the matching; after the matching is completed, the one-to-one matching result of each round of ships and goods is obtained, and the successfully matched goods are removed from the goods set, while the ship set remains unchanged.

8. The market ship intelligent scheduling system based on ship-cargo matching and speed optimization according to claim 6 is characterized in that: In the fitness value calculation and offspring individual generation module, the genetic operators include a selection operator, a crossover operator and a mutation operator; using the genetic operators in the non-dominated sorting genetic algorithm to perform neighborhood search evolution operations on multiple individuals in the initial population to generate multiple new offspring individuals specifically includes: S1: The selection operator selects multiple individuals as parent individuals from the initial population according to the fitness value and adopts the roulette wheel selection method; S2: Determine whether to perform a crossover operation based on the randomly generated probability value and the preset crossover rate. If yes, the crossover operator randomly selects two individuals from the parent individuals, and then randomly selects two speeds with relative positions from the two individuals, and exchanges the codes of these two time units to generate two offspring individuals; if no, the number of parent individuals remains unchanged, and all individuals in the parent individuals are used as offspring individuals; S3: Repeat step S2 until the preset number of iterations is reached to obtain multiple offspring individuals; S4: The mutation operator performs mutation operations on each offspring individual to obtain multiple new offspring individuals.

9. The market ship intelligent scheduling system based on ship-cargo matching and speed optimization according to claim 6 is characterized in that: In the objective function establishment module, the constraints of matching ships with cargo and matching cargo with ships both include a first constraint, a second constraint, a third constraint, a fourth constraint, a fifth constraint, a sixth constraint, a seventh constraint and an eighth constraint. The first constraint is constructed based on the maximum cargo weight, cargo attributes and decision variables of the ship. The second constraint is constructed based on the cargo hold capacity, cargo attributes and decision variables of the ship. The third constraint is constructed based on the maximum draft that the ship can bear, cargo attributes, decision variables, the maximum draft allowed at the location of the cargo and the ship lease status. The fourth, fifth, sixth, seventh and eighth constraints are all constructed based on the decision variables.

10. The market ship intelligent scheduling system based on ship-cargo matching and speed optimization according to claim 9 is characterized in that: In the objective function building module, the ship-cargo constraint condition further includes a ninth constraint condition, and the ninth constraint condition is constructed according to the empty-load sailing mileage and empty-load sailing distance threshold of the ship.

Images