Scheduling optimization method and system based on shipbuilding

Abstract

The invention discloses a scheduling optimization method and system based on shipbuilding, and relates to the technical field of scheduling optimization, and the method specifically comprises the steps: collecting related data of shipbuilding production, and carrying out the preprocessing of the related data; constructing a carrying network based on the preprocessed data, calculating a dependency relationship and a constraint condition between nodes, and generating an optimal block deployment scheme; according to the Gaussian process model, using the optimal block deployment scheme and historical data to construct prior distribution of system behaviors; acquiring real-time state data, and updating and generating posterior probability distribution through the real-time state data; and constructing a mathematical model based on the preprocessed data, and performing multi-objective optimization solution on the mathematical model by adopting an ant colony algorithm and prior distribution or posterior probability distribution of system behaviors to obtain an optimal deployment scheme. According to the method, the Gaussian process and the ant colony algorithm are combined, so that the strategy can be flexibly adjusted to cope with continuously changing conditions, and the scheduling efficiency and precision are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of scheduling optimization, and more particularly to a scheduling optimization method and system based on shipbuilding. Background Art

[0002] At present, as the core pillar of global trade and commodity exchange, the importance of maritime transport is self-evident. The construction and operation efficiency of ships directly affects the smooth operation of the global supply chain. Especially in the current global economic environment, the efficient operation of ships is crucial to maintaining the stability of trade circulation. With the continuous expansion of international trade and the growing demand for ships, the shipbuilding industry has shown a strong development momentum, which has promoted the rapid development of the shipping industry.

[0003] However, the shipbuilding process is complex and diverse, involving three major process flows: hull construction, outfitting, and painting, which include multiple key links, such as steel pretreatment, parts processing, component assembly, section assembly, and slipway assembly. This process faces many challenges, including but not limited to the management of the construction cycle, the effective use of funds, and the coordination between workshop equipment. In particular, existing solutions are insufficient in terms of improving construction efficiency and reducing operating costs. Although deterministic methods can provide optimal solutions for small-scale problems, and heuristic methods provide greater flexibility and scalability, most of these methods are based on static models and lack the dynamic adaptability required to cope with changes in field conditions (such as material shortages, equipment failures, or adjustments to production priorities).

[0004] Therefore, developing a scheduling optimization algorithm with dynamic adaptability to meet the needs of real-time adjustment and emergency response in shipbuilding scenarios is an urgent problem that technical personnel in this field need to solve. Summary of the invention

[0005] In view of this, the present invention provides a scheduling optimization method and system based on shipbuilding, which overcomes the above-mentioned defects.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] A scheduling optimization method based on shipbuilding, the specific steps are:

[0008] Collecting relevant data of shipbuilding production and preprocessing the relevant data;

[0009] Build a loading network based on preprocessed data, calculate the dependencies and constraints between nodes, and generate the optimal overall segment deployment plan;

[0010] Constructing a priori distribution of system behavior using the optimal overall segment deployment plan and historical data according to a Gaussian process model; acquiring real-time status data, and updating and generating a posterior probability distribution through the real-time status data;

[0011] Based on the preprocessed data, a mathematical model is constructed to minimize material handling costs, production cycles and resource consumption and meet multiple constraints. The mathematical model is optimized and solved for multiple objectives using an ant colony algorithm and a priori distribution or a posteriori probability distribution of system behavior to obtain an optimal deployment solution.

[0012] Optionally, the relevant data of ship manufacturing production includes ship design drawings, bill of materials, process flow information, equipment resources and human resource information.

[0013] Optionally, the steps for obtaining the optimal overall segment deployment solution are:

[0014] Building a loading network based on the relevant data, wherein each production link is a node and process constraints are edges in the loading network;

[0015] A topological sorting algorithm is used to calculate the dependencies and constraints between the nodes based on the onboard network to generate the optimal overall segment deployment plan.

[0016] Optionally, the specific steps of the optimal deployment solution are:

[0017] Step 41, construct an optimization model of the shipbuilding process, determine the goal of minimizing material handling costs, production cycle and resource consumption, and define multiple constraints; decompose the shipbuilding process flow into state space nodes, and the state space nodes correspond to production steps, resource allocation and operation procedures;

[0018] Step 42: Initialize the ant colony population, randomly generate multiple ant individuals, each of which represents a shipbuilding path solution, and initialize the initial path of each of the ant individuals;

[0019] Step 43: In each iteration step, the ant individual selects the next state space node based on the path selection probability, and the path selection probability is jointly determined by the pheromone concentration between nodes and the heuristic information;

[0020] Step 44, updating the path pheromone according to the objective function optimization value of the path completed by the individual ant;

[0021] Step 45, evaluate the path plan of each ant individual, select the current optimal solution, and determine whether the iteration termination condition is met; if not, return to step 43 to continue iterative optimization; if so, output the optimal deployment plan.

[0022] Optionally, the expression of the path selection probability is:

[0023] ;

[0024] In the formula, For slave nodes To Node The probability of selection, For the path ( , )’s pheromone strength; For the path ( , ) heuristic information; is the weight of the pheromone; For the path ( , )’s pheromone strength; For the path ( , ) heuristic information; is the weight of the heuristic information; is a candidate path node; A collection of nodes.

[0025] A scheduling optimization system based on shipbuilding includes a data preprocessing module, a deployment sequence optimization module, a heuristic solution generation module and an ant colony optimization module which are connected in sequence;

[0026] The data preprocessing module is used to collect relevant data of shipbuilding production and preprocess the relevant data;

[0027] The deployment sequence optimization module is used to calculate the dependencies and constraints between nodes through the loading network constructed based on the preprocessed data, and generate the optimal overall segment deployment plan;

[0028] The heuristic solution generation module is used to construct a prior distribution of system behavior based on the Gaussian process model using the optimal overall segment deployment solution and historical data; obtain real-time status data, and update and generate a posterior probability distribution through the real-time status data;

[0029] The ant colony optimization module is used to establish a mathematical model that meets multiple constraints with the goal of minimizing material handling costs, production cycles and resource consumption, and uses an ant colony algorithm and a priori distribution or a posteriori probability distribution of system behavior to perform multi-objective optimization on the mathematical model to obtain the optimal deployment plan.

[0030] Optionally, the deployment sequence optimization module includes:

[0031] A loading network construction unit, used to construct a loading network based on the relevant data, wherein each production link is a node and process constraints are edges in the loading network;

[0032] The computing unit is used to calculate the dependencies and constraints between the nodes based on the onboard network using a topological sorting algorithm to generate the optimal overall segment deployment plan.

[0033] Optionally, the ant colony optimization module includes:

[0034] A digital model building unit is used to build an optimization model of the shipbuilding process, determine the goal of minimizing material handling costs, production cycle and resource consumption, and define multiple constraints; decompose the shipbuilding process into state space nodes, and the state space nodes correspond to production steps, resource allocation and operation procedures;

[0035] An ant colony initialization unit, used to initialize the ant colony population, randomly generate a plurality of ant individuals, each of which represents a shipbuilding path solution, and initialize the initial path of each of the ant individuals;

[0036] A path selection unit, configured to select, in each iteration step, the ant individual to select a next state space node based on a path selection probability, wherein the path selection probability is determined by the pheromone concentration between nodes and heuristic information;

[0037] A pheromone updating unit, used for updating the path pheromone according to the objective function optimization value of the path completed by the individual ant;

[0038] The deployment plan generating unit is used to evaluate the path plan of each ant individual, select the current optimal solution, and determine whether the iteration termination condition is met; if not, continue iterative optimization; if so, output the optimal deployment plan.

[0039] It can be seen from the above technical solutions that the present invention discloses a scheduling optimization method and system based on shipbuilding, which has the following beneficial effects compared with the prior art:

[0040] 1. The prior distribution provided by the Gaussian process guides the ant colony algorithm to search, reducing unnecessary exploration and thus reducing the risk of falling into local optimality; as the production environment changes, by updating the posterior distribution, the ant colony algorithm can flexibly adjust its strategy to cope with changing conditions, thereby improving the efficiency and accuracy of scheduling.

[0041] 2. The global search capability of the ant colony algorithm and the local optimization capability of the Gaussian process complement each other. By using the Gaussian process model and heuristic information, the strategy is dynamically adjusted according to historical data and current status, achieving flexible optimization of the production process. In the face of changes and uncertainties in production, the system can respond quickly and adjust strategies, maintaining a high degree of flexibility and adaptability, which is crucial to maintaining the stability and efficiency of the production process.

[0042] 3. Full consideration is given to the various complex constraints encountered in the shipbuilding process, such as safety distance, equipment interference, and task priority, to ensure that the scheduling plan is not only scientific and reasonable, but also feasible, and better adapted to the complex requirements in the actual production environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0044] Figure 1 A schematic diagram of the method provided by the present invention;

[0045] Figure 2 This is a schematic diagram of the system structure provided by the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] On the one hand, an embodiment of the present invention discloses a scheduling optimization method based on shipbuilding, aiming to provide an efficient scheduling optimization method specifically for dealing with the complex scheduling needs in the large-scale shipbuilding industry. The method fully considers the multiple constraints in the shipbuilding process, such as safety distance, interference, and production priority constraints, so that the optimization scheme is more in line with the actual production environment; by combining the Gaussian process and the ant colony algorithm, the accuracy and adaptability are significantly improved. The Gaussian process provides a prior distribution to guide the ant colony algorithm to avoid unnecessary exploration and reduce the risk of falling into local optimality; as the production environment changes and uncertainty increases, the Gaussian process updates the posterior distribution, allowing the ant colony algorithm to flexibly adjust its strategy; in addition, the global search of the ant colony algorithm is combined with the local optimization of the Gaussian process, so that the optimization process has both global exploration capabilities and can finely adjust the optimal solution. This method provides a scientific solution to the yard crane scheduling problem in shipbuilding, and provides a new theoretical basis and practical guidance for optimization problems in complex industrial environments; the steps are as follows Figure 1 As shown, specifically:

[0048] Step 1: Collect relevant data of ship manufacturing and production, and pre-process the relevant data;

[0049] Step 2: Build a loading network based on the preprocessed data, calculate the dependencies and constraints between nodes, and generate the optimal overall segment deployment plan;

[0050] Step 3: Construct a priori distribution of system behavior based on the Gaussian process model using the optimal overall segment deployment plan and historical data; obtain real-time status data, and generate a posterior probability distribution through real-time status data update;

[0051] Step 4: Based on the preprocessed data, a mathematical model is constructed to minimize material handling costs, production cycles, and resource consumption and to meet multiple constraints. The ant colony algorithm and the prior distribution or posterior probability distribution of system behavior are used to perform multi-objective optimization on the mathematical model to obtain the optimal deployment plan.

[0052] In one embodiment, the relevant data of ship manufacturing production includes ship design drawings, bills of materials, process flow information, equipment resources and human resource information.

[0053] Furthermore, data preprocessing is specifically as follows: according to the ship type and manufacturing requirements, the hull is rationally planned and divided into sections; as an intermediate product, the manufacturing process of the section includes small assembly, medium assembly, large assembly, pre-outfitting and painting. Small sections are first combined into medium assemblies, and then combined into large assemblies. After preheating, shot blasting, rust removal and other treatments in the pretreatment workshop, the profiles are transported to the processing area for cutting and bending to form small components. Small components are formed into small components through hoisting, assembly, welding and other processes, and then formed into medium assembly components through similar processes, and finally sent to the pre-outfitting and painting workshops for subsequent processing.

[0054] In one embodiment, the steps for obtaining the optimal overall segment deployment solution are:

[0055] Step 21: construct a loading network based on relevant data, with each production link as a node and process constraints as edges in the loading network;

[0056] Step 22: Use a topological sorting algorithm to calculate the dependencies and constraints between nodes based on the network, and generate an optimal overall segment deployment plan.

[0057] Furthermore, by constructing a loading network, the overall deployment sequence of each production link in the shipbuilding process is optimized. The loading network regards each production link as a node, and generates the optimal deployment sequence by calculating the dependencies and constraints between each node. The optimized deployment sequence helps to reduce conflicts in material handling and equipment scheduling, improve production efficiency, and ensure that each link is carried out in order, thereby effectively shortening the production cycle and reducing production costs.

[0058] Furthermore, the topological sorting algorithm has the following specific calculation steps:

[0059] Initialization: Based on the directed graph of the production process, determine the in-degree of each node (i.e. its predecessor task);

[0060] Select a node: Select a node with an in-degree of zero, indicating that the production link can start;

[0061] Delete nodes and related edges: remove the selected node from the graph and reduce the in-degree of its successor nodes;

[0062] Repeat: Continue selecting nodes with in-degree zero and repeat the above steps until all nodes have been visited.

[0063] In one embodiment, the steps of constructing the Gaussian process model are:

[0064] Select the appropriate kernel function: Select the appropriate Gaussian process kernel function based on historical data to construct the system's prior distribution;

[0065] Training Gaussian process model: Based on historical data, a Gaussian process is used to generate a preliminary prior distribution that reflects the underlying laws and expected behaviors of the shipbuilding process.

[0066] Furthermore, the heuristic information and prior information are used to deploy the total segment; the Gaussian process model is used to generate the prior distribution of system behavior based on historical data, and its posterior distribution is updated in combination with the current state; and the Gaussian process model is used to generate the probability distribution of different deployment schemes. The specific steps are as follows:

[0067] 1) Prior information: Based on historical data (e.g., deployment data from past shipbuilding processes), a prior distribution of system behavior is generated through a Gaussian process, expressed as:

[0068] ;

[0069] In the formula, is the objective function (optimal overall segment deployment plan), is an asymptotic lower bound; is the mean function, is the covariance function, which is used to describe the correlation between different locations.

[0070] 2) Real-time data update: Update the posterior distribution of the Gaussian process model based on the real-time production status (e.g., current process completion status, resource availability, etc.). The posterior distribution is generated based on known training data and current input; the mean expression is:

[0071] ;

[0072] In the formula, Is a new input point The mean of (prior information of training data); Is a new input point and training data The covariance vector between ; is the covariance matrix, is the deviation of the training output from the mean.

[0073] The variance expression is:

[0074] ;

[0075] In the formula, Is a new input point The autocovariance of Is a new input point With training data The covariance between is the inverse of the covariance matrix of the training data; Is a new input point With training data The transposed matrix of the covariance between .

[0076] 3) Generate probability: Generate the probability of optimization schemes (e.g., deployment order, spatial arrangement, etc.) through posterior distribution, so that the deployment order and resource allocation in the shipbuilding process can be predicted and optimized. The expression is:

[0077] ;

[0078] In the formula, is the training data; is the current output; It is a new prediction given historical data; is the uncertainty about the new forecast; is a Gaussian distribution.

[0079] By using Gaussian processes, the effects of various deployment sequences and resource allocation schemes in the shipbuilding process can be predicted based on historical data and current status, and these schemes can be optimized. This method can not only improve the efficiency of scheduling and production, but also handle uncertainty and complex constraints.

[0080] In one embodiment, the specific steps of the optimal deployment solution are:

[0081] Step 41, construct an optimization model of the shipbuilding process, determine the goal of minimizing material handling costs, production cycle and resource consumption, and define multiple constraints; decompose the shipbuilding process flow into state space nodes, and the state space nodes correspond to production steps, resource allocation and operation procedures;

[0082] Step 42: Initialize the ant colony population, randomly generate multiple ant individuals, each ant individual represents a ship manufacturing path plan, and initialize the initial path of each ant individual;

[0083] Step 43, in each iteration step, the individual ant selects the next state space node based on the path selection probability, and the path selection probability is determined by the pheromone concentration between nodes and the heuristic information;

[0084] Step 44, updating the path pheromone according to the objective function optimization value of the path completed by the individual ants;

[0085] Step 45, evaluate the path plan of each ant individual, select the current optimal solution, and determine whether the iteration termination condition is met; if not, return to step 43 to continue iterative optimization; if so, output the optimal deployment plan.

[0086] Furthermore, the site optimization is converted into a mathematical model and optimized through the ant colony algorithm. In the ship site optimization problem, the ant colony algorithm can effectively optimize the key links such as resource allocation and deployment sequence in the hull manufacturing process. The specific steps are:

[0087] In step 41, the problem is modeled: the optimization goal is determined (reducing material handling costs, improving production efficiency, and reducing resource consumption); multiple constraints in shipbuilding are considered (safety distance, equipment interference, job priority, and resource constraints); and the ant colony state space is defined: each step, resource allocation, and production link in the shipbuilding process is used as the state space of the ant colony.

[0088] Material handling costs are quantified based on the distance the material is transported and the number of times it is transported; assuming From the site To site The distance From the site To site The quantity of materials transported, the optimization objective can be expressed as:

[0089] ;

[0090] Production efficiency is usually measured by minimizing the production cycle; the production cycle is the total time to complete the shipbuilding task, which consists of the time of each process. Assume Indicates the process time, the optimization objective can be expressed as:

[0091] ;

[0092] Indicates the number of operations.

[0093] Resource consumption is measured by the consumption of materials, energy, and labor. It is a resource The consumption of It is a resource The unit cost of , the optimization objective can be expressed as:

[0094] .

[0095] In step 42, the ant population is initialized, a certain number of ants are randomly generated, and the path of each ant is initialized. Specifically, each ant represents a potential optimization solution, and the ant will search for the optimal path according to the preset rules. Each ant decides how to choose a path based on some heuristic information (time or space requirements of the production link) and pre-information (historical data or expert knowledge).

[0096] In step 43, in each step, the ant selects the next node to visit (the next production link in the deployment order) according to the path selection rule. This selection is usually based on two factors: pheromone concentration and heuristic information. Each path has a pheromone value, which indicates the quality of the path. The higher the pheromone concentration, the better the path. In shipyard optimization, some heuristic information (such as process time, resource availability, etc.) can be used to help ants make decisions. The expression is usually:

[0097] ;

[0098] In the formula, Represents a slave node To Node The probability of selection; is the path ( , )’s pheromone strength; For the path ( , ) heuristic information; For the path ( , )’s pheromone strength; is the path ( , ) (path feasibility or process completion time), and are the weights of pheromone and heuristic information, respectively; is an index representing a candidate path node; A collection of nodes.

[0099] In step 44, after each ant completes a round of path selection, it will update the pheromone on the path according to the optimization effect of its path (total manufacturing time or resource utilization rate). The pheromone update rule is usually divided into two stages: volatilization and enhancement. The pheromones on all paths will volatilize, simulating the attenuation process of pheromones in nature. If the path taken by an ant is better (i.e., it meets the optimization requirements of the objective function), the pheromone on the path will increase. The pheromone update expression is:

[0100] ;

[0101] In the formula, It is the volatilization factor of pheromone, controlling the speed of pheromone decay; is the path at time t ( , )’s pheromone strength; Ants on the path Increased pheromone amount.

[0102] In step 45, after all ants complete a path selection and pheromone update, they evaluate the quality of each solution. The quality of each path is usually evaluated by an objective function (such as minimizing manufacturing time, resource consumption, etc.). The optimal path is selected as the solution for the current iteration and further optimized based on this solution; steps 43 to 45 are repeated for multiple rounds of iterations until the stop condition is met and the optimal solution that meets the objective function is found.

[0103] Furthermore, ants select paths in each iteration step, and path selection is guided by pheromone intensity and the prior distribution provided by the Gaussian process. The higher the pheromone concentration of the path, the greater the probability that the ant will choose it. The Gaussian process modifies the pheromone intensity through the prior distribution, allowing ants to prioritize those high-probability quality paths. As the uncertainty in the production process increases, the Gaussian process adapts to new environmental changes by updating the posterior distribution in real time. This means that whenever the production environment changes (such as equipment failure, production schedule delays, etc.), the Gaussian process can adjust the posterior distribution to affect the path selection and pheromone update strategy of the ant colony algorithm in real time.

[0104] On the other hand, this embodiment also discloses a scheduling optimization system based on shipbuilding, such as Figure 2 As shown, it includes a data preprocessing module, a deployment sequence optimization module, a heuristic solution generation module and an ant colony optimization module connected in sequence;

[0105] Data preprocessing module, used to collect relevant data of shipbuilding production and preprocess the relevant data;

[0106] The deployment sequence optimization module is used to calculate the dependencies and constraints between nodes through the loading network built based on the preprocessed data, and generate the optimal overall segment deployment plan;

[0107] The heuristic solution generation module is used to construct the prior distribution of system behavior based on the Gaussian process model using the optimal total segment deployment plan and historical data; obtain real-time status data, and generate the posterior probability distribution through real-time status data update;

[0108] The ant colony optimization module is used to establish a mathematical model that meets multiple constraints with the goal of minimizing material handling costs, production cycles, and resource consumption. The ant colony algorithm and the prior distribution or posterior probability distribution of system behavior are used to perform multi-objective optimization on the mathematical model to obtain the optimal deployment plan.

[0109] In one embodiment, the deployment sequence optimization module includes:

[0110] A loading network construction unit is used to construct a loading network based on relevant data, in which each production link is a node and process constraints are edges;

[0111] The computing unit is used to calculate the dependencies and constraints between nodes based on the network using a topological sorting algorithm to generate an optimal overall segment deployment plan.

[0112] In one embodiment, the ant colony optimization module includes:

[0113] The digital model building unit is used to build an optimization model of the shipbuilding process, determine the goal of minimizing material handling costs, production cycle and resource consumption, and define multiple constraints; decompose the shipbuilding process into state space nodes, which correspond to production steps, resource allocation and operation procedures;

[0114] The ant colony initialization unit is used to initialize the ant colony population, randomly generate multiple ant individuals, each ant individual represents a ship manufacturing path plan, and initialize the initial path of each ant individual;

[0115] The path selection unit is used for selecting the next state space node based on the path selection probability of the ant individual in each iteration step. The path selection probability is determined by the pheromone concentration between nodes and the heuristic information.

[0116] A pheromone updating unit, used to update the path pheromone according to the objective function optimization value of the path completed by the individual ants;

[0117] The deployment plan generation unit is used to evaluate the path plan of each ant individual, select the current optimal solution, and determine whether the iteration termination condition is met; if not, continue iterative optimization; if so, output the optimal deployment plan.

[0118] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0119] The above description of the disclosed embodiments enables one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A scheduling optimization method based on shipbuilding, characterized in that: The specific steps are: Collecting relevant data of shipbuilding production and preprocessing the relevant data; Build a loading network based on preprocessed data, calculate the dependencies and constraints between nodes, and generate the optimal overall segment deployment plan; Constructing a prior distribution of system behavior using the optimal overall segment deployment plan and historical data according to a Gaussian process model; Acquire real-time status data, and update and generate a posterior probability distribution through the real-time status data; Based on the preprocessed data, a mathematical model is constructed to minimize material handling costs, production cycles and resource consumption and meet multiple constraints. The mathematical model is optimized and solved for multiple objectives using an ant colony algorithm and a priori distribution or a posteriori probability distribution of system behavior to obtain an optimal deployment solution.

2. A scheduling optimization method based on shipbuilding according to claim 1, characterized in that: The relevant data of ship manufacturing production include ship design drawings, bill of materials, process flow information, equipment resources and human resource information.

3. The scheduling optimization method based on shipbuilding according to claim 1 is characterized in that: The steps for obtaining the optimal overall segment deployment solution are: Building a loading network based on the relevant data, wherein each production link is a node and process constraints are edges in the loading network; A topological sorting algorithm is used to calculate the dependencies and constraints between the nodes based on the onboard network to generate the optimal overall segment deployment plan.

4. A scheduling optimization method based on shipbuilding according to claim 3, characterized in that: The specific steps of the optimal deployment solution are: Step 41, construct an optimization model of the shipbuilding process, determine the goal of minimizing material handling costs, production cycle and resource consumption, and define multiple constraints; decompose the shipbuilding process flow into state space nodes, and the state space nodes correspond to production steps, resource allocation and operation procedures; Step 42: Initialize the ant colony population, randomly generate multiple ant individuals, each of which represents a shipbuilding path solution, and initialize the initial path of each of the ant individuals; Step 43: In each iteration step, the ant individual selects the next state space node based on the path selection probability, and the path selection probability is jointly determined by the pheromone concentration between nodes and the heuristic information; Step 44, updating the path pheromone according to the objective function optimization value of the path completed by the individual ant; Step 45, evaluate the path plan of each ant individual, select the current optimal solution, and determine whether the iteration termination condition is met; if not, return to step 43 to continue iterative optimization; if so, output the optimal deployment plan.

5. A scheduling optimization method based on shipbuilding according to claim 4, characterized in that: The expression of the path selection probability is: ； In the formula, For slave nodes To Node The probability of selection, For the path ( , )’s pheromone strength; For the path ( , ) heuristic information; is the weight of the pheromone; For the path ( , )’s pheromone strength; For the path ( , ) heuristic information; is the weight of the heuristic information; is a candidate path node; A collection of nodes.

6. A scheduling optimization system based on shipbuilding, characterized in that: It includes a data preprocessing module, a deployment sequence optimization module, a heuristic solution generation module and an ant colony optimization module which are connected in sequence; The data preprocessing module is used to collect relevant data of shipbuilding production and preprocess the relevant data; The deployment sequence optimization module is used to calculate the dependencies and constraints between nodes through the loading network constructed based on the preprocessed data, and generate the optimal overall segment deployment plan; The heuristic solution generation module is used to construct a prior distribution of system behavior based on the Gaussian process model using the optimal overall segment deployment solution and historical data; obtain real-time status data, and update and generate a posterior probability distribution through the real-time status data; The ant colony optimization module is used to establish a mathematical model that meets multiple constraints with the goal of minimizing material handling costs, production cycles and resource consumption, and uses an ant colony algorithm and a priori distribution or a posteriori probability distribution of system behavior to perform multi-objective optimization on the mathematical model to obtain the optimal deployment plan.

7. A scheduling optimization system based on shipbuilding according to claim 6, characterized in that: The deployment sequence optimization module includes: A loading network construction unit, used to construct a loading network based on the relevant data, wherein each production link is a node and process constraints are edges in the loading network; The computing unit is used to calculate the dependencies and constraints between the nodes based on the onboard network using a topological sorting algorithm to generate the optimal overall segment deployment plan.

8. A scheduling optimization system based on shipbuilding according to claim 6, characterized in that: The ant colony optimization module includes: A digital model building unit is used to build an optimization model of the shipbuilding process, determine the goal of minimizing material handling costs, production cycle and resource consumption, and define multiple constraints; decompose the shipbuilding process into state space nodes, and the state space nodes correspond to production steps, resource allocation and operation procedures; An ant colony initialization unit, used to initialize the ant colony population, randomly generate a plurality of ant individuals, each of which represents a shipbuilding path solution, and initialize the initial path of each of the ant individuals; A path selection unit, configured to select, in each iteration step, the ant individual to select the next state space node based on a path selection probability, wherein the path selection probability is jointly determined by the pheromone concentration between nodes and heuristic information; A pheromone updating unit, used for updating the path pheromone according to the objective function optimization value of the path completed by the individual ant; The deployment plan generating unit is used to evaluate the path plan of each ant individual, select the current optimal solution, and determine whether the iteration termination condition is met; if not, continue iterative optimization; if so, output the optimal deployment plan.

Images