Watercraft method and apparatus based on genetic algorithm

Abstract

The application provides a water transportation chartering method and device based on a genetic algorithm, and comprises the following steps: predicting the material arrival quantity and the material opening limit quantity of a future date based on a current date and a multi-source data set; determining at least one specified target function from a pre-configured target function set; initializing an order allocation population and a water transportation chartering population; and based on the material arrival quantity and the material opening limit quantity, and by using the specified target function and a constraint set, performing a genetic operation based on co-evolution on the sub-populations included in the order allocation population and the sub-populations included in the water transportation chartering population alternately to obtain a target order allocation plan and a target water transportation chartering plan. The application can significantly improve the automation level of water transportation chartering, improve the efficiency of water transportation chartering, and reduce the labor and time costs consumed by water transportation chartering.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of water transport ship technology, and in particular to a water transport ship method and device based on a genetic algorithm. BACKGROUND

[0002] With the continuous deepening of the digital transformation of the logistics industry, intelligent logistics systems have become the key support for enterprises to improve operational efficiency. Modern logistics management systems have developed from single-function modules to integrated platforms covering warehouse, transportation, distribution and other whole-chain links. The sales logistics module, as the core link connecting the production end and the customer end, directly affects the overall logistics efficiency.

[0003] The sales logistics module in the current intelligent operation integrated system has realized basic informatization management and can centrally collect and analyze data such as inventory status and output plan of multiple production bases. In the water transport delivery link, the system can provide basic data support such as port storage details and order delivery cycle, but the formulation of the ship booking plan still highly depends on manual operation. The specific process is as follows: the logistics officer needs to manually check the ship type demand of different flow directions, manually prepare the ship booking plan in combination with the ship schedule table, and then submit the ship report and coordinate the ship through a third-party logistics service provider. In this mode, manual intervention is required in each link from demand confirmation to ship batch generation, which not only has the risk of inaccurate matching of the ship booking plan and real-time inventory, but also causes response delay due to the speed limit of manual processing, affecting the overall logistics timeliness. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a water transport ship booking method and device based on a genetic algorithm, which can significantly improve the automation level of water transport ship booking, improve the efficiency of water transport ship booking, and reduce the manpower and time cost of water transport ship booking.

[0005] In a first aspect, the present application provides a water transport ship booking method based on a genetic algorithm, comprising:

[0006] Based on the multi-source data set of the current date, predict the material arrival quantity and the material opening limit quantity corresponding to the future date;

[0007] Determine at least one specified target function from the pre-configured target function set;

[0008] Initialize the order allocation population and the water transport ship booking population, use the specified target function and the constraint set, based on the material arrival quantity and the material opening limit quantity, alternately perform the genetic operation based on co-evolution on the sub-population contained in the order allocation population and the sub-population contained in the water transport ship booking population, to obtain the target order allocation plan and the target water transport ship booking plan;

[0009] The target order allocation plan is used for describing to-be-shipped order information corresponding to a future date and target carrier information and target ship information assigned by the to-be-shipped order information, and the target waterway charter plan is used for describing target route information and a target scheduling sequence corresponding to target ship information assigned by to-be-shipped order information in the future date.

[0010] In an implementation, an order allocation population and a waterway charter population are initialized, and based on a material arrival quantity and a material opening limit quantity, a sub-population included in the order allocation population and a sub-population included in the waterway charter population are alternately subjected to a genetic operation based on co-evolution by using a specified objective function and a constraint set, so as to obtain a target order allocation plan and a target waterway charter plan, including:

[0011] The order allocation population includes an order shipping sub-population and an order assignment sub-population, an individual in the order shipping sub-population is combined with an individual in the order assignment sub-population to obtain an order allocation individual, the individual in the order shipping sub-population is used for describing whether a material corresponding to an order is shipped in a future date, and the individual in the order assignment sub-population is used for describing carrier information to which the order is assigned, a ship type, and a corresponding ship quantity.

[0012] The genetic operation based on co-evolution is performed on the sub-population included in the order allocation population by using the specified objective function, so as to obtain a new order allocation population.

[0013] The waterway charter population is initialized based on the new order allocation population, the waterway charter population includes a route allocation sub-population and a scheduling sequence sub-population, an individual in the route allocation sub-population is combined with an individual in the scheduling sequence sub-population to obtain a waterway charter individual, there is at least one waterway charter individual corresponding to one order allocation individual, the individual in the route allocation sub-population is used for describing route information to which a ship is bound, and the individual in the scheduling sequence sub-population is used for describing a corresponding scheduling sequence of the ship.

[0014] The genetic operation based on co-evolution is performed on the sub-population included in the waterway charter population by using the specified objective function, so as to obtain a new waterway charter population.

[0015] The new order allocation population and the new waterway charter population satisfying the constraint set are screened out according to the material arrival quantity and the material opening limit quantity, and the genetic operation based on co-evolution is alternately continued until a first preset iteration stop condition is satisfied, an optimal order allocation individual and a corresponding optimal waterway charter individual are decoded, so as to obtain the target order allocation plan and the target waterway charter plan.

[0016] In an implementation, the genetic operation based on co-evolution is performed on the sub-population included in the order allocation population by using the specified objective function, so as to obtain the new order allocation population, including:

[0017] The order assignment individual is matched with the order shipping individual and the order assignment individual to obtain an order assignment individual;

[0018] A first fitness value corresponding to the order assignment individual is determined by using a specified objective function;

[0019] A plurality of order assignment individuals are screened based on the first fitness value, and genetic operation based on co-evolution is performed on the screened order assignment individuals to obtain new order assignment individuals;

[0020] The first fitness value corresponding to the new order assignment individual is continuously determined until a new order assignment population is obtained when a second preset iteration stop condition is met.

[0021] In an embodiment, the genetic operation based on co-evolution is performed on the screened order assignment individuals to obtain new order assignment individuals, comprising:

[0022] The order shipping individual in the order assignment individual is subjected to genetic operation to obtain a new order shipping individual, and the new order shipping individual is combined with the order assignment individual in the optimal order assignment individual obtained in the previous iteration to obtain a new order assignment individual;

[0023] The order assignment individual in the order assignment individual is subjected to genetic operation to obtain a new order assignment individual, and the new order assignment individual is combined with the order shipping individual in the optimal order assignment individual obtained in the previous iteration to obtain a new order assignment individual;

[0024] A new order assignment population is formed based on the new order assignment individuals.

[0025] In an embodiment, the genetic operation is performed on the order shipping individual in the order assignment individual to obtain a new order shipping individual, comprising:

[0026] It is judged whether the order assignment individual meets a preset condition;

[0027] If yes, the genetic operation is performed on the order shipping individual in the order assignment individual based on a dynamically adjustable memory probability and a genetic operator to obtain a new order shipping individual;

[0028] If no, the genetic operation is performed on the order shipping individual in the order assignment individual based on a preset fixed probability and a genetic operator to obtain a new order shipping individual.

[0029] In an embodiment, the constraint set at least includes a ship load constraint, a logistics balance constraint, an inventory capacity constraint, a consolidation rule, and a capacity limit constraint.

[0030] In an embodiment, the objective function set includes:

[0031] Minimize total warehouse inventory:

[0032] ;

[0033] Minimize number of committed vessels:

[0034] ;

[0035] Minimize weighted order delays:

[0036] ;

[0037] where, is the total warehouse inventory, is the number of committed vessels, is the order delivery delay, is the warehouse consisting of a set of warehouses, is the order consisting of a set of orders, is the order consisting of a set of materials, consisting of a set of carriers, consisting of a set of carriers, consisting of a set of vessel types provided by the carriers, consisting of a set of vessel types provided by the carriers, is a binary variable for an order whether it is shipped within the loading period, is the number of vessels of a vessel type required by an order , consisting of a set of vessel types provided by the carriers, consisting of a set of vessel types provided by the carriers, is a binary variable for an order whether it departs on a future date, is the current inventory of a warehouse , is the weight of a material in an order, is the customer class of an order , is the transportation time after departure of an order , is the delivery date of an order , is the start time of the loading period, is the end time of the loading period.

[0038] ​​​In a second aspect, the present application further provides a water transportation ship arrangement based on a genetic algorithm, comprising:

[0039] a prediction module configured to predict a material arrival quantity and a material departure limit quantity corresponding to a future date based on a multi-source data set of a current date;

[0040] a function determination module configured to determine at least one specified objective function from a pre-configured objective function set;

[0041] a genetic operation module configured to initialize an order allocation population and a water transportation ship arrangement population, and based on the material arrival quantity and the material departure limit quantity, perform a genetic operation based on co-evolution on sub-populations included in the order allocation population and sub-populations included in the water transportation ship arrangement population by using the specified objective function and a constraint set, so as to obtain a target order allocation plan and a target water transportation ship arrangement plan;

[0042] wherein the target order allocation plan is used to describe to-be-shipped order information corresponding to the future date and target carrier information and target ship information assigned by the to-be-shipped order information, and the target water transportation ship arrangement plan is used to describe target route information and a target scheduling sequence corresponding to the target ship information assigned by the to-be-shipped order information.

[0043] In a third aspect, the present application further provides an electronic device, comprising a processor and a memory, the memory stores computer executable instructions capable of being executed by the processor, and the processor executes the computer executable instructions to implement the method provided in any one of the first aspect.

[0044] In a fourth aspect, the present application further provides a computer readable storage medium, the computer readable storage medium stores computer executable instructions, and when the computer executable instructions are called and executed by a processor, the computer executable instructions cause the processor to implement the method provided in any one of the first aspect.

[0045] The application provides a water transportation chartering method and device based on a genetic algorithm, which comprises the following steps: firstly, based on a multi-source data set of a current date, predicting a material arrival quantity and a material opening limit quantity corresponding to a future date; then, determining at least one specified target function from a pre-configured target function set; then, initializing an order allocation population and a water transportation chartering population, and based on the material arrival quantity and the material opening limit quantity, performing a genetic operation based on co-evolution on a sub-population included in the order allocation population and a sub-population included in the water transportation chartering population by using the specified target function and a constraint set, so as to obtain a target order allocation plan and a target water transportation chartering plan; wherein, the target order allocation plan is used for describing to-be-shipped order information corresponding to the future date and target carrier information and target ship information assigned by the to-be-shipped order information, and the target water transportation chartering plan is used for describing target route information and a target scheduling sequence corresponding to the target ship information assigned by the to-be-shipped order information in the future date. The above method firstly predicts the material arrival quantity and the material opening limit quantity corresponding to the future date, and then combines the at least one specified target function, and realizes the joint optimization of the order allocation and the water transportation chartering by alternately performing the genetic algorithm based on co-evolution, so that the automation level of the water transportation chartering can be improved, the efficiency of the water transportation chartering can be improved, and the labor and time cost of the water transportation chartering can be reduced.

[0046] Other features and advantages of the present application will be set forth in the descriptions below, and in part will be apparent from the description, or can be learned by practice of the present application. The purposes and other advantages of the present application will be realized and attained by the structures particularly pointed out in the description, the claims and the drawings.

[0047] In order to make the above objectives, features and advantages of the present application more apparent, the following will describe a preferred embodiment in detail, and the description will be made with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0048] In order to more clearly illustrate the specific embodiments of the present application or the technical solutions in the prior art, the following will briefly introduce the drawings needed to be used in the specific embodiments or the prior art description. Obviously, the drawings described below are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0049] Figure 1 A flowchart of a water transportation chartering method based on a genetic algorithm provided by an embodiment of the present application;

[0050] Figure 2 A schematic diagram of the overall technical framework of a water transportation chartering method based on a genetic algorithm provided by an embodiment of the present application;

[0051] Figure 3A CGA genetic operator schematic diagram of the ship booking plan model provided for the embodiment of the present application;

[0052] Figure 4 A CGA co-evolution strategy principle schematic diagram of the ship booking plan model provided for the embodiment of the present application;

[0053] Figure 5 A structure schematic diagram of the water transport ship booking device based on the genetic algorithm provided for the embodiment of the present application;

[0054] Figure 6 A structure schematic diagram of the electronic device provided for the embodiment of the present application. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical scheme and advantages of the embodiment of the present application clearer, the technical scheme of the present application will be described clearly and completely in combination with the embodiments below. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the present application.

[0056] At present, the water transport delivery process in the sales logistics system needs to be confirmed according to the port storage details, the demand ship type and the ship period of each flow direction, the ship booking plan is prepared, the 3PL is issued by the carrier to report the ship, the ship batch is generated, and then the wharf ship batch information and the expected arrival time are notified. This process needs to consume a large amount of manpower and time cost. In order to improve the delivery efficiency and reduce the finished product inventory in time, a "centralized and consistent" intelligent logistics management and control system needs to be established to enhance the management and control and response ability of the water transport link to ensure low cost, high quality and balanced transportation, and finally to continuously improve the logistics management. Based on this, the present application provides a water transport ship booking method and device based on a genetic algorithm, which can significantly improve the automation level of water transport ship booking, improve the efficiency of water transport ship booking and reduce the manpower and time cost consumed by water transport ship booking.

[0057] In order to facilitate the understanding of the present embodiment, first, a water transport ship booking method based on a genetic algorithm disclosed by the present embodiment will be described in detail, referring to the flowchart of the water transport ship booking method based on a genetic algorithm shown in Figure 1 The method mainly includes the following steps S102 to S106:

[0058] Step S102, based on the multi-source data set of the current date, the material arrival quantity and the material opening limit quantity corresponding to the future date are predicted.

[0059] The multi-source data set can include route information, carrier information, order information, finished product warehouse and port warehouse information, production base information, and transport capacity information. The material arrival quantity at the port refers to the weight of the material arriving at the port on a future date. The material release limit quantity refers to the weight of the material that can be released for transportation on a future date. The material can be steel or the like.

[0060] In step S104, at least one specified objective function is determined from the pre-configured objective function set.

[0061] The objective function set includes minimizing the total warehouse inventory, minimizing the number of reserved ships, and minimizing the weighted order delay. In a specific implementation, at least one objective function can be selected from the minimizing the total warehouse inventory, minimizing the number of reserved ships, and minimizing the weighted order delay as the specified objective function.

[0062] In step S106, the order allocation population and the water transportation reservation population are initialized. Based on the material arrival quantity and the material release limit quantity, the sub-populations included in the order allocation population and the sub-populations included in the water transportation reservation population are alternately subjected to genetic operations based on co-evolution using the specified objective function and the constraint set, to obtain a target order allocation plan and a target water transportation reservation plan.

[0063] The order allocation population includes an order shipment sub-population and an order assignment sub-population. The combination of an individual in the order shipment sub-population and an individual in the order assignment sub-population forms an order allocation individual. The individual in the order shipment sub-population is coded in binary, which is used to describe whether the material corresponding to the order is shipped on a future date. The individual in the order assignment sub-population is coded in natural numbers, which is used to describe the carrier information, the ship type, and the corresponding number of ships to which the order is assigned. The water transportation reservation population includes a route allocation sub-population and a scheduling sequence sub-population. The combination of an individual in the route allocation sub-population and an individual in the scheduling sequence sub-population forms a water transportation reservation individual. One order allocation individual corresponds to at least one water transportation reservation individual. The individual in the route allocation sub-population is coded in binary, which is used to describe the route information of the ship. The individual in the scheduling sequence sub-population is coded in natural numbers, which is used to describe the scheduling sequence of the ship. The constraint set includes ship load constraints, logistics balance constraints, inventory capacity constraints, consolidation rules, release limit constraints, relationship constraints between variables, and variable value constraints. The target order allocation plan is used to describe the to-be-shipped order information on a future date and the target carrier information and target ship information to which the order is assigned. The target water transportation reservation plan is used to describe the target route information and target scheduling sequence corresponding to the target ship information to which the to-be-shipped order information is assigned on a future date.

[0064] In an example, an order allocation population is initialized based on a multi-source data set, a sub-population contained in the order allocation population is subjected to a co-evolution-based genetic operation based on a specified objective function, and then a new order allocation population is obtained; a water charter ship population is initialized based on the new order allocation population, a sub-population contained in the water charter ship population is subjected to a co-evolution-based genetic operation based on a specified objective function, and then a new water charter ship population is obtained; the new order allocation population and the new water charter ship population that satisfy a condition set are screened by using the material arrival quantity and the material limit quantity; the above process is repeated until a first iteration stop (such as reaching a specified number of iterations) condition is met, and then a target order allocation plan and a target water charter ship plan can be obtained based on an optimal order allocation individual and an optimal water charter ship individual corresponding to the optimal order allocation individual.

[0065] The water charter ship method based on the genetic algorithm provided in the embodiments of the present application first predicts the material arrival quantity and the material limit quantity corresponding to a future date, and combines at least one specified objective function to realize the joint optimization of the order allocation and the water charter ship by alternately performing the co-evolution-based genetic algorithm, so that the automation level of the water charter ship can be significantly improved, the efficiency of the water charter ship can be improved, and the human and time costs consumed by the water charter ship can be reduced.

[0066] First, the overall technical framework of the embodiments of the present application is explained and described, referring to Figure 2 The overall technical framework of the water charter ship method based on the genetic algorithm is shown in FIG. 1, which involves a manufacturing end, a sales logistics end, a 3PL (Third-Party logistics, third-party logistics) end, a terminal end, and an e-commerce system end. The manufacturing end determines the material delivery; the sales logistics end performs the following operations: receiving the delivery information, predicting and analyzing the charter model (i.e., the model encapsulated by the water charter ship method based on the genetic algorithm), generating the charter scheme (i.e., the target water charter ship plan), determining whether to adjust, generating the final charter plan if yes, otherwise directly generating the final charter plan, releasing the charter plan, receiving the ship batch performance feedback from the 3PL end, and sending the ship batch information to the 3PL end, the terminal end, and the e-commerce system end; the 3PL end performs the following operations: receiving the final charter plan, and feeding back the ship to the sales logistics end.

[0067] Based on the above overall technical framework, the charter model (i.e., the model encapsulated by the water charter ship method based on the genetic algorithm) is expanded and described. Based on the information of the product type and quantity of the terminal inventory, the production end delivery quantity and the predicted delivery quantity, the required ship type and ship period within a future five-day period are recommended and determined, and the intelligent optimization charter model is constructed to calculate multiple charter schemes for user selection.

[0068] The input, output, constraints, and objectives of the charter model are shown in Table 1 as follows:

[0069] Table 1. Inputs, outputs, constraints, and objectives of the charter model.

[0070]

[0071] The symbol definitions for the ship-to-ship model are shown in Table 2 below:

[0072] Table 2 Symbol Definitions

[0073]

[0074] The objective function of the ship-to-ship model includes:

[0075] Minimize total warehouse inventory:

[0076] ;

[0077] Minimize the number of vessels booked:

[0078] ;

[0079] Minimize weighted order delays (weight: customer level, the higher the customer level, the greater the penalty for delays):

[0080] ;

[0081] in, Total warehouse inventory For the number of ships booked, Due to delays in order delivery, For warehouse The set that is formed For orders The set that is formed For orders Materials included The set that is formed For carriers The set that is formed For carriers Types of ships provided The set that is formed For orders A binary variable indicating whether the shipment occurred within the loading period. For orders Required carriers Types of ships provided The number of ships, For orders Whether in the future date a binary variable, a current inventory of the warehouse, a current inventory of the warehouse, a weight of materials in the order, a customer class of the order, a weight of materials in the order, a customer class of the order, a transportation time after the departure, a delivery date of the order, a transportation time after the departure, a delivery date of the order, a start time of the loading period, an end time of the loading period. a set of constraints, including: ship load constraints, logistics balance constraints, inventory capacity constraints, consolidation rules, open limit constraints, relationship constraints between variables, and variable value constraints. Specifically:

[0082] (1) Ship load constraints: the total weight of finished products in the same batch loaded onto the same ship should not exceed the upper limit of the ship's own load:

[0083]

[0084] ;

[0085] (2) Logistics balance constraints: if the transport ship has collected finished products from the production end, the time period should be greater than or equal to the time when the finished products from the production end arrive at the port:

[0086] ;

[0087] (3) Inventory capacity constraints: the remaining inventory after batching should not be less than the safety inventory of each warehouse:

[0088] ;

[0089] (4) Consolidation rules: the attribute conditions that need to be met for finished products that can be loaded onto the same ship:

[0090] ;

[0091] (5) Open limit constraints:

[0092] ;

[0093] (6) Relationship constraints between variables:

[0094] ;

[0095] ;

[0096] ;​

[0097] ;

[0098] ;

[0099] (7) Variable value constraints:

[0100] ;

[0101] ;

[0102] ;

[0103] It can be seen that the model is a mixed integer nonlinear programming, and the problem solution cannot be obtained in a short time by using mathematical programming and other accurate algorithms, therefore, the intelligent optimization algorithm is used for problem solving.

[0104] Based on this, the implementation process of the embodiment of the application is as follows:

[0105] For the foregoing step S102, the embodiment of the application provides a multi-source data set based on the current date to predict the material arrival quantity and the material opening limit quantity corresponding to the future date.

[0106] Firstly, the existing system is optimized, the information system of the steel plant and the terminal ERP (Enterprise Resource Planning, enterprise resource planning) system are functionally improved and information is integrated, multi-source big data sharing is realized, the ship arrangement, loading distribution and organization direct loading mode are optimized, and manual intervention is reduced.

[0107] Then, according to the manufacturing system, the accurate delivery plan is generated according to the base, the accurate delivery quantity of each flow direction is estimated according to the variety, and the predicted arrival time is generated, so as to generate the material arrival quantity of each flow direction in the future date. Taking the material arrival quantity in the next 5 days as an example, the specific steps are as follows:

[0108] N+5 day arrival quantity: according to the finished product warehouse area, variety and flow direction, the inventory quantity of N-1 day is counted to form the predicted arrival quantity of N+5 day;

[0109] N+4 day arrival quantity: according to the finished product warehouse area, variety and flow direction, the inventory quantity of N-2 day is counted to form the predicted arrival quantity of N+4 day;

[0110] N+3 day arrival quantity: according to the finished product warehouse area, variety and flow direction, the inventory quantity of N-3 day is counted to form the predicted arrival quantity of N+3 day;

[0111] N+2 day arrival quantity: according to the finished product warehouse area, variety and flow direction, the inventory quantity of N-4 day is counted to form the predicted arrival quantity of N+2 day;

[0112] N+1 Day Arrival Volume: The inventory volume of finished products in the warehouse area, product type, and flow direction is calculated to form the expected arrival volume on the N+1 day.

[0113] Every day at midnight, the system automatically generates the material arrival volume for the current day (N+5 days), and uses the data from the previous day (N+2 to N+5) to generate the data for the current day (N+1 to N+4).

[0114] Next, the system can obtain the approved shipment quantity and the shipment entrustment quantity according to the contract sub-item. Based on the entrustment quantity and the quantity in port, the opening limit quantity is calculated. There is no opening limit mark on the material, and the material is not bound to the shipment entrustment. It is impossible to know the specific opening limit time of the material. However, the vessel allocation time and the container loading plan time can be calculated. Based on the vessel allocation time and the container loading plan time, the opening limit rules are formulated and automatically matched to each material, thereby predicting the material opening limit quantity within the next 5 days.

[0115] Finally, based on the flow statistics, a statistical table of inventory and material release limits for each flow direction is generated. By using the material release limits and material arrival volumes, the estimated shipment volume for the next 5 days can be predicted, according to the following rules:

[0116] N+5 Day Booking Quantity: The material supply limit for N days, plus the material arrival quantity (supply limit) for the next 5 days, forms the booking quantity for the N+5th day;

[0117] N+4 Day Booking Quantity: The material supply limit for N days, plus the material arrival quantity (supply limit) for the next 4 days, forms the booking quantity for the N+1 day;

[0118] N+3 Day Shipment Quota: The material supply limit for N days, plus the material arrival volume (supply limit) for the next 3 days, forms the shipment quota for the N+1 day;

[0119] N+2 Day Shipment Quota: The material supply limit for N days, plus the material arrival volume (supply limit) for the next 2 days, forms the shipment quota for the N+1 day;

[0120] N+1 Day Shipment Quota: The material supply limit for N days, plus the material arrival limit for the next day, forms the shipment quota for the N+1 day.

[0121] For the foregoing step S106, based on the quantity in port, the quantity in transit of each production base, and the forecast quantity of finished product warehouse of each production base, and with the experience of business personnel as an aid, the system will automatically call an artificial intelligence algorithm based on the quantity in port, the quantity in transit of each production base, and the forecast quantity of finished product warehouse of each production base, the flow direction and delivery period of the finished product contract, and other information, the available ship type, the load capacity, and the constraints of the finished product chartering rules, to achieve the principle of minimizing the cost of finished product inventory and ship logistics transportation under the premise of meeting the constraints of ship load capacity and warehouse capacity, to decide the ship type and ship period required for the daily finished product shipment, to generate multiple target water transportation chartering plans, and to display the obtained target water transportation chartering plan only in the form of a table. The operation personnel can manually modify the target water transportation chartering plan automatically generated by the computer according to the influence of uncertain factors, recommend multiple target water transportation chartering plans to business personnel through data integration and model prediction, reduce the labor intensity of business personnel, improve work efficiency, realize intelligent recommendation of cargo control and chartering, and optimize chartering.

[0122] The embodiment of the present application introduces a co-evolutionary genetic algorithm (Co-evolutionary GA, CGA), provides a method for initializing an order allocation population and a water transportation chartering population, uses a specified objective function and a constraint set, and based on the material quantity in port and the material opening limit, alternately performs co-evolutionary-based genetic operations on the sub-populations contained in the order allocation population and the sub-populations contained in the water transportation chartering population to obtain a target order allocation plan and a target water transportation chartering plan, including:

[0123] (1) Initialize the order allocation population. The order allocation population contains an order shipment sub-population and an order assignment sub-population. The combination of individuals in the order shipment sub-population and individuals in the order assignment sub-population obtains an order allocation individual. The individual in the order shipment sub-population is used to describe whether the material corresponding to the order is shipped in the future date, and the individual in the order assignment sub-population is used to describe the carrier information to which the order is assigned, the ship type, and the corresponding number of ships.

[0124] The order shipment sub-population is a binary code, indicating the shipment of the order in the loading period. If the order is shipped in the loading period, then the order is not shipped temporarily.

[0125] The order assignment sub-population is a natural number matrix code, indicating the number of b type ships of carrier c to which the order is assigned.

[0126] In the initialization of the order allocation population, it is first determined whether the population size meets the requirements. If the population size is set too small, it will lead to insufficient samples in the genetic operation process, increasing the risk of premature convergence. If the population size is too large, it will increase the calculation cost. The appropriate population size needs to be determined according to the specific complexity of the ship scheduling model and the number of decision variables. For example, for small and medium-sized ship scheduling problems, the population size can be set to 50-100; for large and complex problems, the population size can be appropriately increased to about 200. On this basis, the order shipment sub-population X and the order assignment sub-population Y are randomly generated, and the two sub-populations are combined to form a complete order allocation population.

[0127] (2) Based on the specified objective function, the sub-populations contained in the order allocation population are subjected to genetic operation based on co-evolution, to obtain a new order allocation population.

[0128] Co-evolution divides the complex multi-decision optimization problem into multiple sub-problems (subsets), and each subset corresponds to a type of decision variable. This way, each sub-population can evolve independently, avoiding the complex interference problems that may occur when all decision variables are mixed.

[0129] In addition, the embodiments of the present application design genetic operators for each sub-population according to its coding characteristics, such as Figure 3 As shown in a CGA genetic operator diagram of a ship scheduling plan model, the genetic operator of the order shipment sub-population X: Since the order shipment sub-population X is binary coded, the genetic operator can include binary crossover (such as single-point crossover, multi-point crossover) and binary mutation (such as bit flip mutation), which are specifically designed for the characteristics of binary coding and can efficiently explore the search space; the genetic operator of the order assignment sub-population Y: For the natural number coded order assignment sub-population Y, the genetic operator can use numerical based operations, such as integer crossover (such as arithmetic crossover) and integer mutation (such as adding or subtracting a random integer). This design can better adapt to the characteristics of natural number coding and the constraints of the problem.

[0130] Further, the embodiments of the present application design a dynamic probability parameter design method based on memory, to realize adaptive updating of probability values.

[0131] On this basis, the specific process of generating a new order allocation population through genetic operation based on co-evolution is as follows:

[0132] (2.1) Match the order shipping individuals and the order assignment individuals contained in the order allocation population to obtain order allocation individuals. For example, randomly combine the order shipping individuals and the order assignment individuals to obtain an order allocation individual.

[0133] (2.2) Determine the first fitness value corresponding to the order allocation individual by using a specified objective function. In an example, at least one objective function is selected as the specified objective function based on actual demand from minimizing the total amount of warehouse inventory, minimizing the number of reserved ships, and minimizing the weighted order delay, and then the specified objective function is used as a fitness function to evaluate the first fitness value of the order allocation individual. The fitness function is usually set according to the objective of the ship chartering model. For example, if the objective is to minimize the total cost, the fitness value can be the inverse of the total cost, and the higher the fitness, the better the scheme. If the objective is to maximize the benefit, the fitness value can be the benefit value directly.

[0134] (2.3) Screen a plurality of order allocation individuals based on the first fitness value, and perform genetic operation based on co-evolution on the screened order allocation individuals to obtain new order allocation individuals. Specifically.

[0135] In an example, the process of screening the order allocation individuals is as follows: find the individual with the highest fitness value from the order allocation individuals as the current optimal individual For example, assuming that the total cost of individual 1 is 500, the total cost of individual 2 is 480, and the total cost of individual 3 is 520, the fitness value of individual 2 is the highest, and individual 2 will be recorded as the current optimal individual.

[0136] In an example, referring to the principle diagram of the CGA co-evolution strategy of the ship chartering plan model shown in Figure 4 , each iteration first performs genetic operation on the population of the sub-population X to obtain a set of offspring , combines the sub-population X in with the current optimal subset to form an individual set and evaluates and selects to obtain a new population ; then, the same method is used to evolve the population of the sub-population Y, and thus a complete evolution of one generation is completed. In the cycle process of co-evolution, as long as the second preset iteration stopping condition (such as reaching the maximum number of iterations, fitness convergence, etc.) is not met, the evolution operation will continue. For example, the maximum number of iterations is set to 100, and when the algorithm runs to the 100th iteration, the evolution process is terminated. Based on the above principle, the process of genetic operation based on co-evolution is as follows:

[0137] (I) Genetic operation is performed on the order delivery individual in the order allocation individual to obtain a new order delivery individual, and the new order delivery individual is combined with the order assignment individual in the optimal order allocation individual obtained in the previous iteration to obtain a new order allocation individual.

[0138] In an example, it is determined whether the order allocation individual satisfies a preset condition. The preset condition can be whether the similarity between the current order allocation individual and the historical order allocation individual is greater than a threshold value. If yes, it is determined that the preset condition is satisfied, and genetic operation is performed on the order delivery individual in the order allocation individual based on the dynamically adjustable memory probability and the genetic operator to obtain a new order delivery individual. If no, it is determined that the preset condition is not satisfied, and genetic operation is performed on the order delivery individual in the order allocation individual based on the preset fixed probability and the genetic operator to obtain a new order delivery individual.

[0139] The specific implementation process is as follows: it is determined according to certain conditions whether to generate a child set according to the memory probability and call the genetic operator (crossover operator) and (mutation operator). The memory probability reflects the influence of the past successful experience of the genetic operator on the current operation. If the condition is satisfied, the crossover and mutation operations are guided by the memory probability and. For example, the crossover and mutation operators with higher memory probability and are preferentially selected, so that the past successful experience can be learned and the search efficiency can be improved. If the condition is not satisfied, the two operators are called according to the specified probability to generate the child set. For example, the crossover probability is specified as 0.8 and the mutation probability is specified as 0.2, so that the crossover and mutation operations are performed according to the probabilities when generating the child. At the same time, the cumulative execution times of the two operators are updated, which are used for subsequent calculation of their success rates and other indicators.

[0140] Further, the memory probability is updated and the sum is normalized, so that these probability values can reasonably guide the use of the genetic operator in the subsequent evolution process. At the same time, the optimal subpopulation is updated, that is, the fitness of the subpopulation X in the current child is compared, and the subpopulation X with the highest fitness is retained as the new optimal subpopulation .

[0141] (I) Genetic operation is performed on the order delivery individual in the order allocation individual to obtain a new order delivery individual, and the new order delivery individual is combined with the order assignment individual in the optimal order allocation individual obtained in the previous iteration to obtain a new order allocation individual.

[0142] Similarly, similar to the evolution operation of the subpopulation X, it is first determined whether to generate a child set according to the memory probability and call (crossover operator), (insertion operator) and If the condition is satisfied, the offspring set is generated by calling the three kinds of operators in turn, and if the condition is not satisfied, the offspring is generated by calling the three kinds of operators according to the specified probability, and the cumulative execution times of the three kinds of operators are updated. The memory probability and the optimal sub-population are updated by calling the memory rules 1-3 and the summation normalization method That is, the fitness of the sub-population Y in the current offspring is compared, and the sub-population Y with the highest fitness is retained as the new optimal sub-population .

[0143] (III) Based on the new order allocation individual, a new order allocation population is formed.

[0144] Specifically, the sub-population X and the sub-population Y are decoded, that is, the binary code and the natural number code of the subset are converted into actual decision variable values, such as converting the binary vector of the sub-population X into the shipping situation of the order, and converting the natural number vector of the sub-population Y into the number of each type of ship. Then, the parent-child tournament method is called, and the parent and child individuals are selected according to the evaluation rules (such as the fitness value), the individuals with higher fitness are retained into the next generation, and the cumulative success times of the genetic operators are updated, so that the quality of the population can be gradually improved, and the subsequent use probability of the genetic operator can be adjusted according to the success of the genetic operator.

[0145] (2.4) Continue to determine the first fitness value corresponding to the new order allocation individual until the second preset iteration stopping condition is met to obtain the new order allocation population.

[0146] (3) Based on the new order allocation population, the water transport charter population is initialized. The water transport charter population includes a route allocation sub-population and a scheduling order sub-population, the individual in the route allocation sub-population is combined with the individual in the scheduling order sub-population to obtain a water transport charter individual, there is at least one water transport charter individual corresponding to one order allocation individual, the individual in the route allocation sub-population is used to describe the route information of the bound ship, and the individual in the scheduling order sub-population is used to describe the scheduling order corresponding to the ship.

[0147] In one embodiment, for a new order assignment population, at least one route assignment individual and one scheduling order individual can be randomly generated for any order assignment individual, and the combination of the two is the water transport contract ship individual corresponding to the order assignment individual. For example, the route assignment sub-population can be represented as a binary vector, and if the ship needs to perform a specific task (such as direct transportation of goods, direct / transfer combined transportation of goods), the planning result will clearly indicate the route to which each ship is assigned. For example, whether a ship is assigned to a specific route (such as binary representation: 1 means assignment, 0 means no assignment), such as [1, 0, 1, 0], which means the first ship is assigned to route 1, the second ship is not assigned to route 1, the third ship is assigned to route 1, and the fourth ship is not assigned to route 1. The scheduling order sub-population can be a natural number vector, which is used to represent the scheduling order of the ship, for example, where and when each ship starts and completes the task, such as [3, 1, 4, 2], which means the ships are scheduled in the order of ship No. 3, ship No. 1, ship No. 4, and ship No. 2. The purpose of randomly generating these subsets is to provide diverse initial solutions for the algorithm and increase the scope of the search.

[0148] In another embodiment, the water transport contract ship population can also include a terminal assignment sub-population, a route assignment sub-population, and a scheduling order sub-population, wherein the terminal assignment sub-population and the route assignment sub-population are both binary coded, and the scheduling order sub-population is natural number coded. The terminal assignment sub-population is to determine the specific terminal where each ship docks. For example, a total of 10 ships need to be docked at 5 terminals, and the individual will contain the terminal number to which each ship is assigned.

[0149] On this basis, based on the scheduling order individual, the value of the variable can be determined.

[0150] (4) Based on the specified objective function, perform co-evolution-based genetic operations on the sub-populations included in the water transport contract ship population to obtain a new water transport contract ship population. The co-evolution-based genetic operations can refer to the aforementioned (2.1) to (2.4), and the embodiments of the present application do not repeat them. In addition, during the co-evolution-based genetic operation process, the water transport contract ship individual can be screened according to the following constraints: it cannot be less than the minimum contract amount of the ship, the contract plan cannot conflict with the existing ship, special tasks need to be completed within a specific time, etc.

[0151] For example, assuming there are 3 ships (Nos. 1, 2, and 3), 2 terminals (A and B), and 2 routes (X and Y). A possible water transport contract ship individual can be represented as follows:

[0152] Ship assignment:

[0153] Ship 1: assigned to terminal A, not assigned to route X, assigned to route Y.

[0154] Ship 2: assigned to terminal B, assigned to route X, not assigned to route Y.

[0155] Ship 3: assigned to terminal A, assigned to route X, not assigned to route Y.

[0156] The individual in the approximate ship model is a specific embodiment of the planning result, which covers the allocation of ships, time scheduling, and the satisfaction of resource and constraint conditions, etc. By optimizing these contents, the algorithm can find a better ship scheduling scheme to achieve the optimization goal of minimizing cost or maximizing benefit, etc.

[0157] (5) According to the material arrival quantity and the material opening limit quantity, a new order allocation population and a new water transport ship population that meet the constraint set are screened out, and the genetic operation based on co-evolution is continued to be alternately performed until the first preset iteration stopping condition is met, and the optimal order allocation individual and the optimal water transport ship individual corresponding thereto are decoded to obtain a target order allocation plan and a target water transport ship plan.

[0158] Among them, the material opening limit quantity specifically refers to the opening limit quantity of the material in the port According to the material opening limit quantity Individuals meeting the opening limit quantity constraint can be screened out, and individuals meeting the ship load constraint, the logistics balance constraint, the inventory capacity constraint, the consolidation rule, the relationship constraint between variables, and the variable value constraint can be further screened out to form a new order allocation population and a new water transport ship population.

[0159] The foregoing (1) to (5) are repeated, and when the first preset iteration stopping condition (such as the number of iterations or convergence) is met, the order allocation individual with the highest first fitness value is taken as the optimal order allocation individual, the water transport ship individual with the second fitness value (that is, the fitness value obtained in the process of genetic operation based on co-evolution on the water transport ship population) is taken as the optimal water transport ship individual, and both are decoded into actual decision variable values to obtain a target order allocation plan and a target water transport ship plan. The scheme should be the optimal or better ship plan found in the iteration process of the algorithm, which can better meet the goals of the ship model, such as the lowest cost or the maximum benefit, etc.

[0160] In summary, the water transport ship method based on genetic algorithm provided by the embodiment of the application has at least the following characteristics: automatic: replacing manual calculation of ship allocation and allocation. Intelligent: using intelligent decision-making methods to realize artificial replacement. Optimization: exploring various potential to realize plan optimization, allocation optimization, and freight saving. Collaboration: multi-system linkage and seamless collaboration.

[0161] On the basis of the foregoing embodiments, the embodiments of the present application provide a water transportation chartering device based on a genetic algorithm, referring to Figure 5 A structural schematic diagram of a water transportation chartering device based on a genetic algorithm is shown in the figure, and the device mainly comprises the following parts:

[0162] A prediction module 502 is configured to predict the material arrival quantity and the material opening limit quantity corresponding to a future date based on a multi-source data set of the current date;

[0163] A function determination module 504 is configured to determine at least one specified target function from a pre-configured target function set;

[0164] A genetic operation module 506 is configured to initialize an order allocation population and a water transportation chartering population, and based on the material arrival quantity and the material opening limit quantity, perform collaborative evolution-based genetic operations on the sub-populations contained in the order allocation population and the sub-populations contained in the water transportation chartering population alternately by using the specified target function and the constraint set, so as to obtain a target order allocation plan and a target water transportation chartering plan;

[0165] The target order allocation plan is used to describe the to-be-shipped order information corresponding to the future date and the target carrier information and the target ship information assigned by the to-be-shipped order information, and the target water transportation chartering plan is used to describe the target route information and the target scheduling order corresponding to the target ship information assigned by the to-be-shipped order information.

[0166] The water transportation chartering device based on the genetic algorithm provided by the embodiments of the present application firstly predicts the material arrival quantity and the material opening limit quantity corresponding to a future date, and combines at least one specified target function, and realizes the joint optimization of the order allocation and the water transportation chartering by alternately performing the collaborative evolution-based genetic algorithm, so as to significantly improve the automation level of the water transportation chartering, improve the efficiency of the water transportation chartering, and reduce the manpower and time cost consumed by the water transportation chartering.

[0167] In an implementation manner, the genetic operation module 506 is specifically configured to:

[0168] initialize the order allocation population; wherein the order allocation population comprises an order shipping sub-population and an order assignment sub-population, a combination of an individual in the order shipping sub-population and an individual in the order assignment sub-population obtains an order allocation individual, the individual in the order shipping sub-population is used to describe whether the material corresponding to an order is shipped in a future date, and the individual in the order assignment sub-population is used to describe the carrier information to which the order is assigned, the ship type, and the corresponding ship quantity;

[0169] perform collaborative evolution-based genetic operations on the sub-populations contained in the order allocation population based on the specified target function, to obtain a new order allocation population;

[0170] initialize a water charter fleet based on the new order allocation population; wherein the water charter fleet comprises a route allocation sub-population and a dispatch order sub-population, an individual in the route allocation sub-population is combined with an individual in the dispatch order sub-population to obtain a water charter individual, at least one water charter individual corresponds to one order allocation individual, the individual in the route allocation sub-population is used to describe the bound route information of the ship, and the individual in the dispatch order sub-population is used to describe the corresponding dispatch order of the ship;

[0171] perform a genetic operation on the sub-population included in the water charter fleet based on the specified target function to obtain a new water charter fleet;

[0172] filter the new order allocation population and the new water charter fleet that meet the constraint set according to the material arrival quantity and the material opening limit, and continue to alternately perform the genetic operation based on the co-evolution until a first preset iteration stop condition is met, and then decode the optimal order allocation individual and the corresponding optimal water charter individual to obtain a target order allocation plan and a target water charter plan.

[0173] In an embodiment, the genetic operation module 506 is specifically configured to:

[0174] match the order shipping individual and the order assignment individual included in the order allocation population to obtain an order allocation individual;

[0175] determine a first fitness value corresponding to the order allocation individual by using the specified target function;

[0176] filter a plurality of order allocation individuals based on the first fitness value, and perform a genetic operation on the filtered order allocation individuals based on the co-evolution to obtain new order allocation individuals;

[0177] continue to determine the first fitness value corresponding to the new order allocation individual until a second preset iteration stop condition is met to obtain a new order allocation population.

[0178] In an embodiment, the genetic operation module 506 is specifically configured to:

[0179] perform a genetic operation on the order shipping individual in the order allocation individual to obtain a new order shipping individual, combine the new order shipping individual with the order assignment individual in the optimal order allocation individual obtained in the previous iteration to obtain a new order allocation individual;

[0180] and perform a genetic operation on the order allocation individual in the order allocation individual to obtain a new order allocation individual, and combine the new order allocation individual with the order shipping individual in the optimal order allocation individual obtained in the previous iteration to obtain a new order allocation individual;

[0181] A new order allocation population is formed based on the individuals assigned to the new orders.

[0182] In one implementation, the genetic operation module 506 is specifically used for:

[0183] Determine whether the individual order allocation meets the preset conditions;

[0184] If so, then based on the dynamically adjustable memory probability and genetic operators, genetic operations are performed on the order shipping individuals in the order allocation individuals to obtain new order shipping individuals;

[0185] If not, then based on the preset fixed probability and genetic operator, perform genetic operations on the order dispatch individuals in the order allocation individuals to obtain new order dispatch individuals.

[0186] In one implementation, the constraint set includes at least: ship load constraints, logistics balance constraints, inventory capacity constraints, distribution rules, and opening limit constraints.

[0187] In one implementation, the set of objective functions includes:

[0188] Minimize total warehouse inventory:

[0189] ;

[0190] Minimize the number of vessels booked:

[0191] ;

[0192] Minimize weighted order delays:

[0193] ;

[0194] in, Total warehouse inventory For the number of ships booked, Due to delays in order delivery, For warehouse The set that is formed For orders The set that is formed For orders Materials included The set that is formed For carriers The set that is formed For carriers Types of ships provided The set that is formed For orders A binary variable indicating whether the shipment occurred within the loading period. For orders Required carrier Type of vessel offered Number of vessels, For order Whether in the future date Binary variable of departure, For warehouse Current inventory, For order Weight of materials For order Customer rating of order Transportation time after departure, Delivery date of order For order For the start time of the loading period, For the end time of the loading period. The device provided by the embodiment of the application has the same implementation principle and generated technical effects as the foregoing method embodiment. For brevity of description, the part not mentioned in the device embodiment can be referred to the corresponding content in the foregoing method embodiment.

[0195] The embodiment of the application provides an electronic device, and specifically, the electronic device comprises a processor and a storage device; the storage device stores a computer program, and the computer program performs the method according to any one of the foregoing embodiments when the computer program is run by the processor.

[0196]

[0197] Figure 6 The electronic device 100 provided by the embodiment of the application comprises a processor 60, a memory 61, a bus 62 and a communication interface 63, the processor 60, the communication interface 63 and the memory 61 are connected through the bus 62; the processor 60 is configured to execute an executable module stored in the memory 61, for example, a computer program.

[0198] The memory 61 can include a high-speed random access memory (RAM) and can also include a non-volatile memory, for example, at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 63 (which can be wired or wireless), and the Internet, a wide area network, a local area network, a metropolitan area network and the like can be used.

[0199] The bus 62 can be an ISA bus, a PCI bus or an EISA bus and the like. The bus can be divided into an address bus, a data bus, a control bus and the like. For the convenience of representation,​​Figure 6 Only one bus or type of bus might exist but there is a possibility of the existence of more buses or more types of buses.

[0200] The memory 61 is configured to store a program, and the processor 60 executes the program after receiving an execution instruction. The method executed by the device for defining a flow process according to any of the foregoing embodiments of the application can be applied to the processor 60 or implemented by the processor 60.

[0201] The processor 60 can be an integrated circuit chip having a processing capability of signals. In the implementation process, each step of the foregoing method can be completed by an integrated logic circuit of hardware in the processor 60 or an instruction in the form of software. The foregoing processor 60 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), and the like; or can be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component. Each method, step and logic block diagram disclosed in the embodiments of the application can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in combination with the embodiments of the application can be directly embodied as a hardware code processor for execution, or a combination of hardware and software modules in the code processor for execution. The software module can be located in a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, or other mature storage medium in the art. The storage medium is located in the memory 61, and the processor 60 reads information in the memory 61 and combines the hardware to complete the steps of the foregoing method.

[0202] The computer program product of the readable storage medium provided by the embodiments of the application includes a computer readable storage medium storing a program code, and the program code includes instructions for executing the method described in the foregoing method embodiments. The specific implementation can be referred to the foregoing method embodiments, and will not be described here.

[0203] If the functions are realized in the form of software function units and sold or used as independent products, they can be stored in a computer readable storage medium. Based on this understanding, the technical solutions of the present application or the parts of the present application that essentially contribute to the prior art or the parts of the technical solutions can be embodied in the form of software products. The computer software product is stored in a storage medium and includes instructions for making a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in the embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0204] Finally, it should be noted that: the above-described embodiments are only specific embodiments of the present application, which are used to illustrate the technical solutions of the present application, but not to limit them. The protection scope of the present application is not limited to this. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily think of changes to the technical solutions recorded in the foregoing embodiments within the technical range disclosed by the present application, or make equivalent replacements to some technical features. The modifications, changes or replacements do not make the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for booking ships in water transport based on a genetic algorithm, characterized in that, include: Based on the multi-source data set of the current date, predict the material arrival volume and material release limit for the corresponding future date. The material release limit refers to the weight of material that can be released for transportation on the future date. Determine at least one specified objective function from a pre-configured set of objective functions; Initialize the order allocation population and the shipping booking population. Using the specified objective function and constraint set, based on the material arrival quantity and the material opening limit, alternately perform co-evolutionary genetic operations on the subpopulations of the order allocation population and the subpopulations of the shipping booking population to obtain the target order allocation plan and the target shipping booking plan. This includes: initializing the order allocation population; performing co-evolutionary genetic operations on the subpopulations of the order allocation population based on the specified objective function to obtain a new order allocation population; and based on the new order allocation population, initially... Initialize the shipping vessel population; based on the specified objective function, perform genetic operations based on co-evolution on the subpopulations included in the shipping vessel population to obtain a new shipping vessel population; according to the material arrival quantity and the material opening limit, select a new order allocation population and a new shipping vessel population that satisfy the constraint set, and continue to perform genetic operations based on co-evolution alternately until the first preset iteration stopping condition is met, then decode the optimal order allocation individual and its corresponding optimal shipping vessel individual to obtain the target order allocation plan and the target shipping vessel plan; The target order allocation plan describes the pending shipment order information and the assigned target carrier and target vessel information for a future date. The target waterway booking plan describes the target route information and target scheduling order corresponding to the target vessel information assigned to the pending shipment order information for the future date.

2. The method for booking ships in water transport based on a genetic algorithm according to claim 1, characterized in that, The order allocation population includes an order shipment subpopulation and an order assignment subpopulation. Individuals in the order shipment subpopulation and individuals in the order assignment subpopulation are combined to obtain an order allocation individual. Individuals in the order shipment subpopulation are used to describe whether the materials corresponding to the order will be shipped on the future date. Individuals in the order assignment subpopulation are used to describe the carrier information, vessel type and corresponding number of vessels to which the order is assigned. The water transport vessel booking population includes a route allocation sub-population and a scheduling order sub-population. Individuals in the route allocation sub-population and individuals in the scheduling order sub-population are combined to obtain water transport vessel booking individuals. One order allocation individual corresponds to at least one water transport vessel booking individual. Individuals in the route allocation sub-population are used to describe the route information bound to the vessel, and individuals in the scheduling order sub-population are used to describe the scheduling order corresponding to the vessel.

3. The method for booking ships in water transport based on a genetic algorithm according to claim 2, characterized in that, Based on the specified objective function, genetic operations based on co-evolution are performed on the subpopulations included in the order allocation population to obtain a new order allocation population, including: The order dispatch individuals and order assignment individuals contained in the order allocation population are matched to obtain the order allocation individuals; Using the specified objective function, determine the first fitness value corresponding to the individual assigned the order; Multiple order allocation individuals are selected based on the first fitness value, and genetic operations based on co-evolution are performed on the selected order allocation individuals to obtain new order allocation individuals; Continue to determine the first fitness value corresponding to the new order allocation individual until the second preset iteration stopping condition is met to obtain the new order allocation population.

4. The method for booking ships in water transport based on a genetic algorithm according to claim 3, characterized in that, Perform co-evolution-based genetic operations on the selected order allocation individuals to obtain new order allocation individuals, including: Genetic operations are performed on the order shipping individuals in the order allocation individuals to obtain new order shipping individuals. The new order shipping individuals are then combined with the order assignment individuals in the optimal order allocation individuals obtained in the previous iteration to obtain new order allocation individuals. In addition, a genetic operation is performed on the order allocation individuals to obtain new order allocation individuals, and the new order allocation individuals are combined with the order shipping individuals in the optimal order allocation individuals obtained in the previous iteration to obtain new order allocation individuals; A new order allocation population is formed based on the newly assigned individuals.

5. The method for booking ships in water transport based on a genetic algorithm according to claim 4, characterized in that, Genetic operations are performed on the order shipment individuals in the order allocation individuals to obtain new order shipment individuals, including: Determine whether the individual assigned to the order meets the preset conditions; If so, then based on the dynamically adjustable memory probability and genetic operator, a genetic operation is performed on the order dispatch individual in the order allocation individual to obtain a new order dispatch individual; If not, then based on the preset fixed probability and the genetic operator, a genetic operation is performed on the order dispatch individual in the order allocation individual to obtain a new order dispatch individual.

6. The method for booking ships in water transport based on a genetic algorithm according to claim 1, characterized in that, The constraint set includes at least: ship load constraints, logistics balance constraints, inventory capacity constraints, distribution rules, and opening limit constraints.

7. The method for booking ships in water transport based on a genetic algorithm according to claim 1, characterized in that, The set of objective functions includes: Minimize total warehouse inventory: ； Minimize the number of vessels booked: ； Minimize weighted order delays: ； in, Total warehouse inventory For the number of ships booked, Due to delays in order delivery, For warehouse The set that is formed For orders The set that is formed For orders Materials included The set that is formed For carriers The set that is formed For carriers Types of ships provided The set that is formed For orders A binary variable indicating whether the shipment occurred within the loading period. For orders Required carriers Types of ships provided The number of ships, For orders Whether in the future date The binary variable for departure. For warehouse Current inventory levels For orders medium material The weight, For orders Customer level, For orders Transportation time after departure For orders Delivery date, This is the start time of the loading period. This is the end time of the loading period.

8. A water transport vessel booking device based on a genetic algorithm, characterized in that, include: The prediction module is used to predict the material arrival quantity and material release limit for future dates based on a multi-source data set of the current date. The material release limit refers to the weight of material that can be released for transportation on a future date. The function determination module is used to determine at least one specified objective function from a pre-configured set of objective functions; The genetic operation module is used to initialize the order allocation population and the shipping booking population. Using the specified objective function and constraint set, and based on the material arrival quantity and the material opening limit, it alternately performs co-evolution-based genetic operations on the subpopulations of the order allocation population and the subpopulations of the shipping booking population to obtain the target order allocation plan and the target shipping booking plan. This includes: initializing the order allocation population; performing co-evolution-based genetic operations on the subpopulations of the order allocation population based on the specified objective function to obtain a new order allocation population; and performing genetic operations based on the new order allocation population... The process involves: 1) mating the population and initializing the water transport vessel booking population; 2) performing co-evolution-based genetic operations on the subpopulations of the water transport vessel booking population based on the specified objective function to obtain a new water transport vessel booking population; 3) selecting new order allocation populations and new water transport vessel booking populations that satisfy the constraint set based on the material arrival quantity and the material opening limit quantity, and continuing to perform co-evolution-based genetic operations alternately until the first preset iteration stopping condition is met; 4) decoding the optimal order allocation individual and its corresponding optimal water transport vessel booking individual to obtain the target order allocation plan and the target water transport vessel booking plan. The target order allocation plan describes the pending shipment order information and the assigned target carrier and target vessel information for a future date. The target waterway booking plan describes the target route information and target scheduling order corresponding to the target vessel information assigned to the pending shipment order information for the future date.

9. An electronic device, characterized in that, The method includes a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the method of any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions that, when invoked and executed by a processor, cause the processor to perform the method according to any one of claims 1 to 7.

Images