Fetal position scheduling method, device and equipment for ship construction

Abstract

The invention relates to the technical field of ship construction, and provides a tire position scheduling method, device and equipment for ship construction, and the method comprises the steps: obtaining ship segment data and tire position attribute data; according to the ship segment data and the tire position attribute data, constructing a scheduling constraint condition and a scheduling optimization function; generating a plurality of initial scheduling strategies according to scheduling constraint conditions in combination with the subsection overdue risk priority of each ship subsection; performing gene coding on the initial scheduling strategy to obtain an initial gene sequence; performing iterative genetic operation on the initial gene sequence by adopting a non-dominated sorting genetic algorithm based on the scheduling optimization function to obtain a Pareto optimal solution set; and performing gene decoding on the Pareto optimal solution set, and determining a fetal position scheduling strategy. According to the technical scheme provided by one or more embodiments of the invention, a reasonable tire position scheduling scheme can be automatically generated, and the global optimality and the actual suitability of the tire position scheduling scheme can be ensured.

Description

Technical Field

[0001] This disclosure relates to the field of shipbuilding technology, specifically to a method, apparatus, and equipment for positioning in shipbuilding. Background Technology

[0002] The modern shipbuilding industry commonly adopts a modular, sectional construction method. By breaking down the entire ship into multiple blocks and manufacturing and assembling them in parallel, construction efficiency can be greatly improved. In mass shipbuilding, the sectional construction station is the "workstation" of the sectional construction assembly line, serving as the work space and equipment for operations such as panel laying, assembly, and welding.

[0003] However, the capacity of space allocation is relatively small and the demand is concentrated in the entire production process, making it a core bottleneck resource in the production system. Therefore, the rationality of space allocation scheduling is of great significance for shortening the shipbuilding cycle, improving shipbuilding efficiency, and reducing shipbuilding costs. In actual production, most existing space allocation scheduling methods used for ship section construction adopt manual scheduling, which relies heavily on the production experience of factory personnel and is difficult to handle complex constraints. This inevitably leads to explicit or implicit conflicts in the scheduling results, causing production rhythm disruptions.

[0004] Furthermore, some shipyards have attempted to adopt automated tire position scheduling methods. However, most existing automated scheduling algorithms lack solutions to address the multi-objective conflicts inherent in tire position scheduling. During optimization, they may overlook diverse value demands in actual production, leading to solutions that lack rationality and practicality, and causing production rhythm disruptions. In addition, existing automated scheduling algorithms are prone to getting trapped in local optima, failing to fully exploit the potential of tire position resources and limiting the maximization of production capacity. Summary of the Invention

[0005] In view of this, one or more embodiments of this disclosure provide a method, apparatus and equipment for scaffolding in ship construction, which can automatically generate reasonable scaffolding schemes and ensure the global optimality and practical adaptability of the scaffolding schemes. It is suitable for ship section construction production environments with limited resources and multiple tasks in parallel.

[0006] In a first aspect, this disclosure provides a method for scheduling positions in ship construction. The method includes: acquiring ship segment data and position attribute data, wherein the ship segment data characterizes the expected attributes of each ship segment; constructing scheduling constraints and a scheduling optimization function based on the ship segment data and the position attribute data, wherein the scheduling constraints include spatial constraints, functional constraints, and temporal constraints, and the scheduling optimization function includes total completion time, total overdue days, total number of overdue segments, and the difference in completion time between sub-segments within the same parent segment; generating multiple initial scheduling strategies based on the scheduling constraints and the segment overdue risk priority of each ship segment; encoding the initial scheduling strategies to obtain initial gene sequences, wherein each initial gene sequence corresponds to a position distribution scheme, and the gene arrangement position in the initial gene sequence corresponds one-to-one with the segment index; performing iterative genetic operations on the initial gene sequences using a non-dominated sorting genetic algorithm based on the scheduling optimization function to obtain a Pareto optimal solution set; and decoding the Pareto optimal solution set to determine the position scheduling strategy.

[0007] Secondly, this disclosure provides a tire position scheduling device for ship construction. The device includes: a data input unit for acquiring ship segment data and tire position attribute data, wherein the ship segment data characterizes the expected attributes of each ship segment; a data modeling unit for constructing scheduling constraints and a scheduling optimization function based on the ship segment data and the tire position attribute data, wherein the scheduling constraints include spatial constraints, functional constraints, and time constraints, and the scheduling optimization function includes total completion time, total overdue days, total overdue segments, and the difference in completion time between sub-segments within the same parent segment; and an initial scheduling unit for... The system generates multiple initial scheduling strategies based on the scheduling constraints and the segment overdue risk priority of each ship segment. A gene encoding unit encodes the initial scheduling strategies to obtain initial gene sequences, each corresponding to a fetal position distribution scheme. The gene arrangement positions in the initial gene sequences correspond one-to-one with the segment indices. An optimized scheduling unit performs iterative genetic operations on the initial gene sequences using a non-dominated sorting genetic algorithm based on the scheduling optimization function to obtain a Pareto optimal solution set. A gene decoding unit decodes the Pareto optimal solution set to determine the fetal position scheduling strategy.

[0008] Thirdly, this disclosure provides an electronic device including a memory and a processor. The memory stores a computer program, which, when executed by the processor, implements the above-described method for tire positioning in ship construction.

[0009] Fourthly, this disclosure provides a computer-readable storage medium for storing a computer program that, when executed by a processor, implements the above-described method for tire position scheduling in ship construction.

[0010] This disclosure provides a technical solution through one or more embodiments, constructing a precise mathematical model for tire placement scheduling that aligns with shipyard production efficiency and risk management needs based on ship section data and tire position attribute data. It also establishes accurate scheduling constraints and optimization functions. The scheduling constraints, including spatial, functional, and temporal constraints, fully consider and mitigate various explicit or implicit conflicts, ensuring the order of ship production. The scheduling optimization function, designed in conjunction with actual ship production needs, can comprehensively optimize four target directions: total completion time, total overdue days, total overdue sections, and the difference in completion time between sub-sections within the same parent section, increasing the rationality and practicality of the tire placement scheduling strategy. The initial scheduling strategy, obtained by combining scheduling constraints with the priority of section overdue risks, ensures both the spatial and functional requirements of each ship section for tire positions and a low overdue section rate, laying a solid foundation for subsequent algorithm optimization.

[0011] This disclosure provides a technical solution through one or more embodiments, encoding the initial scheduling strategy into an initial gene sequence. This not only efficiently indicates the specific scheduling processing task for each fetal position (which ship segment to process at what time), but also facilitates the effective execution of the non-dominated sorting genetic algorithm. Based on the scheduling optimization function, after each iteration of the genetic operation, the non-dominated sorting genetic algorithm can comprehensively evaluate the optimization effect of the current fetal position scheduling strategy in multiple target directions, retaining a subset of fetal position scheduling strategies with better overall performance. After multiple iterations of genetic operations, the non-dominated sorting genetic algorithm can output a Pareto optimal solution set, which can then be decoded to obtain a fetal position scheduling strategy that takes into account multiple target directions. Attached Figure Description

[0012] The features and advantages of the embodiments of this disclosure will be more clearly understood by referring to the accompanying drawings, which are illustrative and should not be construed as limiting the present disclosure in any way. In the drawings: Figure 1 A schematic diagram illustrating the steps of a tire position scheduling method for shipbuilding according to one embodiment of this disclosure is shown; Figure 2 A schematic diagram of a gene sequence and its decoding result is shown in one embodiment of this disclosure; Figure 3 A schematic diagram of a non-dominant sorting method is shown in one embodiment of this disclosure; Figure 4 A schematic diagram of a crossover mutation operation is shown in one embodiment of this disclosure; Figure 5 This diagram illustrates the convergence of total completion days in one embodiment of the present disclosure. Figure 6 This diagram illustrates the total overdue segmented convergence in one embodiment of the present disclosure; Figure 7 This diagram illustrates the convergence of total overdue segment time in one embodiment of the present disclosure. Figure 8 This diagram illustrates the convergence of the completion time difference of the same parent segment in one embodiment of this disclosure. Figure 9 This invention provides a schematic diagram comparing the performance indicators of different algorithms obtained in one embodiment of the present disclosure. Figure 10 A schematic diagram of the functional modules of a tire position scheduling device for shipbuilding according to one embodiment of the present disclosure is shown. Figure 11 A schematic diagram of the structure of an electronic device according to one embodiment of the present disclosure is shown. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0014] Please see Figure 1 The present disclosure provides a method for scheduling the position of a ship during construction, which may include the following steps.

[0015] S1: Obtain ship section data and tire position attribute data, wherein the ship section data represents the expected attributes of each ship section.

[0016] In this embodiment, based on ship section data and tire position attribute data, a precise tire position scheduling mathematical model that meets the shipyard's production efficiency and risk management needs can be constructed, and accurate scheduling constraints and scheduling optimization functions can be formulated.

[0017] In a practical application example, ship section data can be represented as follows: . It can represent the first i The expected attribute data for each ship section, that is, the attribute data of the ship section during construction and after completion. Among them, and Indicates the length and width of a ship's sections. and This indicates the processing time and completion deadline for each ship section. This indicates the parent section to which a ship section belongs (some sections belong to the same parent section). For construction building assembly B subset of ), This indicates the available construction superstructure for this vessel section. The tire position attribute data is represented as follows: It consists of attribute data of all fetal positions distributed within different construction buildings. s The attribute data of each fetal position can be represented as: .in, and Indicates the length and width of the fetal position. The building where the fetus is located is indicated. This refers to the number of construction buildings. It should be noted that the processing procedures for each jig within the same construction building are the same.

[0018] S2: Based on the ship segment data and the tire position attribute data, construct scheduling constraints and scheduling optimization functions. The scheduling constraints include spatial constraints, functional constraints, and time constraints. The scheduling optimization functions include total completion time, total overdue days, total overdue segments, and the difference in completion time between sub-segments within the same parent segment.

[0019] In this embodiment, racking scheduling is a resource scheduling optimization problem with strong constraints. Due to the significant differences in structure and technology among different ship sections, a specific section can only be processed in one or a few racks equipped with the corresponding process equipment. Furthermore, since section construction follows strict process logic, the completion of some sections is a prerequisite for the commencement of subsequent sections, and this dependency constitutes a strict timing constraint.

[0020] In this implementation, scheduling constraints, including spatial, functional, and temporal constraints, fully consider and mitigate various explicit or implicit conflicts, ensuring the order of ship production. For example, spatially, the jacquard space allocated to a section must be able to accommodate the corresponding section; a jacquard space generally can only accommodate one section; special sections may require the simultaneous use of two adjacent jacquard spaces; and when multiple sections use the same jacquard space, the maximum capacity of the jacquard space must not be exceeded. Functionally, the jacquard space allocated to a section must meet the section's operational requirements. Temporally, the time spent by different sections in the same jacquard space cannot overlap.

[0021] In this implementation, the segmented construction scheduling in actual production needs to consider multiple standards, such as minimizing the completion time, minimizing the number of overdue days, and minimizing the completion interval of sub-segments within the same parent segment, which correspond to shorter production cycles, lower delivery delay risks, and more compact production rhythms, respectively.

[0022] In this embodiment, the scheduling optimization function designed in accordance with the actual needs of ship production can comprehensively optimize four types of target directions: total completion time, total overdue days, total overdue segments, and the difference in completion time of sub-segments within the same parent segment, thereby increasing the rationality and practicality of the tire position scheduling strategy.

[0023] In some implementations, the ship segment data includes segment length, segment width, segment processing cycle, segment completion deadline, parent segment to which the segment belongs, and optional construction decks for the segment; the tire position attribute data includes tire position length, tire position width, and the construction deck where the tire position is located; the construction of scheduling constraints and scheduling optimization functions includes: constructing the functional constraints such that the construction deck where the target tire position is located belongs to the optional construction decks of the target segment, and the target tire position is a candidate for allocation to the target segment; constructing the spatial constraints such that the segment length of the target segment is less than the tire position length of the target tire position, and the segment width of the target segment is less than the tire position width of the target tire position; and constructing the time constraints such that the scheduling processing periods of multiple target segments on the same target tire position do not overlap.

[0024] In a practical application example, referring to the ship section data and tire position attribute data in the previous practical application example, the specific formula for the scheduling constraints can be as follows.

[0025]

[0026]

[0027]

[0028]

[0029]

[0030] Among them, the i The scheduling results of each segment are used express, This indicates the assigned fetal position. This indicates the start time of processing. This indicates the segment index that shares the fetal position with it. Assigned to fetal position. i The task sequence can be represented as The first formula can be expressed as a piecewise division. iThe assigned workstation space must be located within a construction building that meets the processing requirements. Formulas two through four represent the spatial constraints when one segment occupies one workstation, multiple segments share one workstation, and one segment occupies multiple workstation spaces, respectively. Formula five indicates that different segment tasks assigned to the same workstation must not overlap in time.

[0031] In a practical application example, the construction of scheduling constraints and scheduling optimization functions includes: constructing a first optimization function corresponding to the total completion time. Construct a second optimization function corresponding to the total number of overdue days. Construct a third optimization function corresponding to the total number of overdue segments. Construct a fourth optimization function corresponding to the completion time difference of sub-segments within the same parent segment. ;in, Indicates the segmented processing cycle. Indicates the deadline for completion of each segment. Indicates the total number of segments. Indicates the start time of the segment processing. This indicates the parent segment to which the segment belongs.

[0032] S3: Based on the scheduling constraints and the segment overdue risk priority of each ship segment, generate multiple initial scheduling strategies.

[0033] In this implementation, the initial scheduling strategy obtained by combining scheduling constraints and segment overdue risk priority ensures that each ship segment meets the space and functional requirements of the space available for the space, while also ensuring a low overdue segment rate, thus laying a solid foundation for subsequent optimization of the algorithm.

[0034] In some implementations, the step of generating multiple initial scheduling strategies based on the scheduling constraints and the segment delinquency risk priority of each vessel segment includes: determining a preset number of segment scheduling time periods based on the tire position attribute data; calculating the latest start time for processing each vessel segment based on the segment processing cycle and segment completion deadline; determining the segment delinquency risk priority of each vessel segment based on the latest start time for processing each vessel segment and the tire position scheduling time periods; and scheduling each vessel segment according to the segment delinquency risk priority based on the scheduling constraints to generate the initial scheduling strategy.

[0035] In some implementations, scheduling each vessel segment according to the segment overdue risk priority includes: if there are segments of the same level with the same overdue risk priority, then the segments of the same level are scheduled in a random order.

[0036] Specifically, this disclosure adopts an initialization strategy based on segmented overdue risk priority. First, the fetal position scheduling time interval is divided into N time periods, and the latest start time for each segment is calculated based on the segment processing days and the latest delivery time. Then, based on the time period in which the latest start time of each segment falls, the segments are divided into N segmented overdue risk priorities. Next, the segments are scheduled according to their segmented overdue risk priorities, with segments at the same priority being scheduled randomly. While ensuring that the fetal positions and times of the scheduled segments remain unchanged, the earliest available fetal position that satisfies spatial and functional constraints is selected for the next scheduled segment. Combining segmented overdue risk priorities to generate several initial scheduling strategies can accelerate the convergence of the non-dominated sorting genetic algorithm and ensure the engineering feasibility of the algorithm results.

[0037] S4: Genetically encode the initial scheduling strategy to obtain an initial gene sequence. Each initial gene sequence corresponds to a fetal position distribution scheme, and the gene arrangement position in the initial gene sequence corresponds one-to-one with the segment index.

[0038] In this embodiment, by using a gene encoding and decoding method adapted to the fetal position scheduling scenario, the initial scheduling strategy can be encoded into an initial gene sequence, which not only efficiently indicates the specific scheduling processing task for each fetal position, but also facilitates the effective execution of the non-dominated sorting genetic algorithm.

[0039] In this embodiment, please refer to Figure 2 Each gene in the gene sequence represents a segmented task, and the order of the genes in the sequence represents the segment index. Optionally, by designing a three-dimensional coding structure of "construction building number - fetal position number - scheduling priority", the scheduling information of the corresponding segment can be further clearly represented. Based on the gene sequence, the construction building and fetal position number of each segment can be directly determined. After decoding the gene sequence, the specific processing time of each segment in a specific fetal position can be obtained.

[0040] In a practical application example, after decoding the gene sequence, a list of all processing tasks assigned to each fetal position can be extracted (the fetal position of each segment can be directly obtained from the gene information). Then, the segment tasks assigned to the same fetal position are scheduled according to priority, and the specific processing time for each segment can be determined. Through this process, the gene sequence is mapped into a complete scheduling scheme that includes the fetal position for processing each segment and the processing start time.

[0041] S5: Based on the scheduling optimization function, a non-dominated sorting genetic algorithm is used to perform iterative genetic operations on the initial gene sequence to obtain the Pareto optimal solution set.

[0042] In this embodiment, non-dominated sorting genetic algorithms (such as NSGA-II) possess strong global search capabilities and fast convergence speed. Using the gene encoding sequences of the scheduling constraints, scheduling optimization functions, and initial scheduling strategies determined in the preceding steps of this embodiment, a non-dominated sorting genetic algorithm can be employed to solve for the Pareto optimal solution set. The Pareto optimal solution set output after multiple iterations effectively improves the global optimality and practical adaptability of the scheduling scheme, making it suitable for resource-constrained, multi-task parallel ship section construction production environments.

[0043] In this embodiment, the number of iterations in the iterative genetic operation can be flexibly set according to the actual application scenario and is not specifically limited. Optionally, based on historical experience, the number of iterations can be preferentially set between 80 and 200 to balance computational accuracy and computational efficiency. Optionally, a specific numerical value for the number of iterations can be not preset, and after each round of iterative genetic operation, the numerical results of one or more scheduling optimization functions can be statistically analyzed as reference values; if the reference values ​​determined by multiple rounds of iterative genetic operation tend to stabilize (for example, determined by whether the rate of change is within a preset range), the iterative genetic operation can be automatically stopped.

[0044] In this implementation, based on the scheduling optimization function, the non-dominated sorting genetic algorithm can comprehensively evaluate the optimization effect of the current fetal position scheduling strategy in multiple target directions after each iteration of genetic operations, retaining a subset of fetal position scheduling strategies with better overall performance. For example, a four-dimensional fitness function composed of total completion time, weighted total overdue time, number of overdue segments, and delivery time difference of segments within the same mother is used to evaluate the performance of the fetal position scheduling strategy. New individual gene sequences are generated generation by generation through iterative operations such as non-dominated sorting, crossover, and mutation, retaining the best-performing scheduling scheme.

[0045] In some embodiments, the step of performing iterative genetic operations on the initial gene sequence using a non-dominated sorting genetic algorithm based on the scheduling optimization function includes: determining the non-dominated sorting of the initial gene sequence based on the calculation result of the scheduling optimization function; calculating the distribution density of the initial gene sequences in the same non-dominated layer according to the non-dominated sorting, and determining the crowding degree of the initial gene sequences in the same non-dominated layer; selecting the initial gene sequence with a high non-dominated level to obtain an iterative gene sequence, which is used to update and replace the initial gene sequence in the next round of iterative genetic operations; and selecting the initial gene sequence with a high crowding degree among the initial gene sequences in the same non-dominated layer to obtain the iterative gene sequence.

[0046] Specifically, non-dominated sorting and crowding calculation are the core steps that distinguish non-dominated sorting genetic algorithms from traditional genetic algorithms. The "dominance" relationship is a fundamental rule for comparing the merits of two solutions in multi-objective optimization. For solutions A and B, if A dominates B, then solution A is no worse than B on all optimization objectives (in this scenario, total completion time, total overdue days, number of overdue segments, and difference in completion time between sub-segments within the same parent segment), and solution A is superior to B on at least one objective. For example... Figure 3 As shown, Scheme 1 dominates Schemes 4, 5, and 6, while Schemes 1, 2, and 3 have a non-dominant relationship. Non-dominant sorting stratifies all solutions in the population: first, all individuals not dominated by any other solution are identified and marked as the first non-dominant layer (Pareto front). Then, new non-dominant solutions are identified from the remaining solutions as the second layer, and this process is repeated until all solutions are assigned to a certain layer. Following this principle, solutions in higher layers have a "dominant relationship" with solutions in lower layers (the higher layer solution is absolutely superior to the lower layer solution), while solutions in the same layer have a "non-dominant relationship" (it is impossible to determine which is superior).

[0047] "Crowding" is used to measure the distribution density of solutions in the same non-dominated layer. For each objective function, the solutions in that layer are first sorted according to the objective value (each solution corresponds to a four-dimensional integer sequence), and then the objective space distance of each solution is calculated. The larger the distance, the greater the crowding, indicating that the solution is more spacious and has a better distribution.

[0048] When selecting the next generation of the genetic algorithm, non-dominated sorting prioritizes individuals with high non-dominated levels and, within the same set, prioritizes individuals with high crowding. This design ensures the stable retention of high-quality solutions and, by preserving sparsely distributed individuals, effectively prevents the population from clustering in localized regions. This ensures the algorithm can continuously explore different regions of the solution space and significantly reduces the risk of getting trapped in local optima.

[0049] In some implementations, the step of using a non-dominated sorting genetic algorithm to perform iterative genetic operations on the initial gene sequence based on the scheduling optimization function further includes: for the iterative gene sequence, adding assigned weights to the gene loci corresponding to the overdue segments, performing weighted random selection, determining the crossover site, and exchanging the gene loci in the two iterative gene sequences that match the crossover site to generate a new iterative gene sequence.

[0050] Specifically, crossover sites can be randomly generated within a sequence length range. To promote the improvement of low-quality gene fragments, overdue segment selection has a higher probability of crossover when generating crossover sites. Then, the genes at the crossover sites of the two parent individuals are exchanged to generate offspring individuals; see [link to relevant documentation]. Figure 4As shown, this mechanism enhances gene diversity while preserving the high-quality features of the solution, and provides a greater recombination probability for segments with larger delays, thus achieving a good balance between the ability to explore the optimal solution and the convergence speed of the algorithm.

[0051] In some implementations, the step of using a non-dominated sorting genetic algorithm to perform iterative genetic operations on the initial gene sequence based on the scheduling optimization function further includes: for the iterative gene sequence, adding assigned weights to the gene loci corresponding to the overdue segments, performing weighted random selection, determining mutation sites, and exchanging gene loci in one of the iterative gene sequences that match the mutation sites to generate a new iterative gene sequence.

[0052] Specifically, mutation operations can also employ differentiated mutation probabilities. Lower mutation probabilities are used for general genes, while higher mutation probabilities are used for gene loci corresponding to overdue segments. For example... Figure 4 As shown, the mutation process involves adjusting the position or priority of segments. Position adjustments must meet the space and processing function requirements of the segment, while priority adjustments involve appropriately increasing the priority based on the original value. Through this mutation strategy, the algorithm can centrally optimize the scheduling of overdue segments while avoiding changes to the overall scheme, thus ensuring the stability of the scheduling scheme.

[0053] S6: Perform gene decoding on the Pareto optimal solution set to determine the fetal position scheduling strategy.

[0054] In this embodiment, the non-dominated sorting genetic algorithm, after multiple iterative genetic operations, can output a Pareto optimal solution set. The Pareto optimal solution set is a core concept in multi-objective optimization, referring to the set of all solutions for which "no single objective can be improved without worsening other objectives." In other words, after genetic decoding of the Pareto optimal solution set, multiple fetal position scheduling strategies can be determined. Theoretically, none of these strategies is superior to another. At this point, any one of these fetal position scheduling strategies chosen by the user is the optimal strategy that balances multiple objectives. Optionally, the user can select from these fetal position scheduling strategies based on their own preferences (e.g., the minimum total overdue days, the minimum number of overdue segments, etc.).

[0055] In some implementations, after determining the fetal position scheduling strategy, the method further includes: receiving and responding to a user input instruction, and outputting a preferred scheduling strategy in the fetal position scheduling strategy that matches the user input instruction.

[0056] Specifically, users can input selection commands through the displayed interface of the fetal position scheduling strategy. Users can also directly input option commands, such as "minimum total overdue days" or "minimum total completion time".

[0057] This disclosure provides a method for scheduling hull sections in shipbuilding, and its effectiveness in this area can be verified through a specific simulation experiment. The experiment selected 514 hull section processing tasks from a shipyard as the test set, with 193 hull sections of varying sizes and processing functions as the hull section resources. Using total completion time, total overdue time, number of overdue sections, and the difference in completion time between sections within the same parent section as evaluation indicators, the hull section scheduling method provided in this disclosure was compared with traditional automated scheduling algorithms based on genetic algorithms (GA) or multi-objective particle swarm optimization (MOPSO). All experiments were run under the same hardware conditions, with an initial population (initial scheduling strategy) size of 25 and 150 iterations.

[0058] Experimental results are as follows Figures 5-9 As shown, Figures 5-8 The historical best convergence results of the four objective functions under different algorithms are shown respectively. Figure 5 (Convergence status of total completion days) Figure 6 (Overdue segmented convergence status) Figure 7 (Convergence status of total overdue periods in segments) Figure 8 (Convergence of completion time difference for the same parent segment) shows the historical best convergence of the four objective functions under different algorithms. It can be seen that both MOPSO and the genetic algorithm show an overall decreasing trend, but their efficiency is far lower than that of the NSGA-II algorithm. The genetic algorithm's convergence curve converges prematurely because, for single-objective optimization, the genetic algorithm lacks a process for comparing multiple optimal solutions when dealing with multi-objectives, thus prematurely getting trapped in local optima. Figure 9 The comparison of different algorithms across four metrics is presented. The NSGA-II algorithm demonstrates its superiority and efficiency by having the fewest overdue segments, the fewest overdue days, and the fewest average overdue days.

[0059] It should be noted that the main focus of this disclosure is on how to construct scheduling constraints and optimization functions that precisely meet the production efficiency and risk management needs of shipyards, how to generate initial scheduling strategies, the encoding and decoding logic of the initial scheduling strategies and gene sequences, and how to combine these with a non-dominated sorting genetic algorithm to obtain a place-based scheduling strategy. The algorithmic steps of the non-dominated sorting genetic algorithm itself are secondary aspects of this disclosure. For example, based on the scheduling constraints, optimization functions, and initial scheduling strategies provided in this disclosure, other multi-objective optimization algorithms can also achieve more superior and practical place-based scheduling schemes compared to existing technologies.

[0060] This disclosure provides a technical solution through one or more embodiments, constructing a precise mathematical model for tire placement scheduling that aligns with shipyard production efficiency and risk management needs based on ship section data and tire position attribute data. It also establishes accurate scheduling constraints and optimization functions. The scheduling constraints, including spatial, functional, and temporal constraints, fully consider and mitigate various explicit or implicit conflicts, ensuring the order of ship production. The scheduling optimization function, designed in conjunction with actual ship production needs, can comprehensively optimize four target directions: total completion time, total overdue days, total overdue sections, and the difference in completion time between sub-sections within the same parent section, increasing the rationality and practicality of the tire placement scheduling strategy. The initial scheduling strategy, obtained by combining scheduling constraints with the priority of section overdue risks, ensures both the spatial and functional requirements of each ship section for tire positions and a low overdue section rate, laying a solid foundation for subsequent algorithm optimization.

[0061] This disclosure provides a technical solution through one or more embodiments, encoding the initial scheduling strategy into an initial gene sequence. This not only efficiently indicates the specific scheduling processing task for each fetal position (which ship segment to process at what time), but also facilitates the effective execution of the non-dominated sorting genetic algorithm. Based on the scheduling optimization function, after each iteration of the genetic operation, the non-dominated sorting genetic algorithm can comprehensively evaluate the optimization effect of the current fetal position scheduling strategy in multiple target directions, retaining a subset of fetal position scheduling strategies with better overall performance. After multiple iterations of genetic operations, the non-dominated sorting genetic algorithm can output a Pareto optimal solution set, which can then be decoded to obtain a fetal position scheduling strategy that takes into account multiple target directions.

[0062] Please see Figure 10 This disclosure also provides a tire position scheduling device for shipbuilding, the device comprising: The data input unit 100 is used to acquire ship segment data and tire position attribute data, wherein the ship segment data represents the expected attributes of each ship segment; The data modeling unit 200 is used to construct scheduling constraints and scheduling optimization functions based on the ship segment data and the tire position attribute data. The scheduling constraints include spatial constraints, functional constraints and time constraints. The scheduling optimization function includes total completion time, total overdue days, total overdue segments and the difference in completion time of sub-segments within the same parent segment. The initial scheduling unit 300 is used to generate multiple initial scheduling strategies based on the scheduling constraints and the segment overdue risk priority of each ship segment. Gene coding unit 400 is used to encode the initial scheduling strategy to obtain an initial gene sequence. Each initial gene sequence corresponds to a fetal position distribution scheme. The gene arrangement position in the initial gene sequence corresponds one-to-one with the segment index. The optimization scheduling unit 500 is used to perform iterative genetic operations on the initial gene sequence based on the scheduling optimization function and using a non-dominated sorting genetic algorithm to obtain a Pareto optimal solution set. Gene decoding unit 600 is used to perform gene decoding on the Pareto optimal solution set to determine the fetal position scheduling strategy.

[0063] In one embodiment, the ship segment data includes segment length, segment width, segment processing cycle, segment completion deadline, parent segment to which the segment belongs, and optional construction towers for the segment. The tire position attribute data includes tire position length, tire position width, and the construction tower where the tire position is located. The data modeling unit 200 is specifically used to: construct the functional constraints such that the construction tower where the target tire position is located belongs to the optional construction tower of the target segment, and the target tire position is a candidate allocated to the target segment; construct the spatial constraints such that the segment length of the target segment is less than the tire position length of the target tire position, and the segment width of the target segment is less than the tire position width of the target tire position; and construct the time constraints such that the scheduling processing time periods of multiple target segments on the same target tire position do not overlap.

[0064] In one implementation, the data modeling unit 200 is further configured to: construct a first optimization function corresponding to the total completion time. Construct a second optimization function corresponding to the total number of overdue days. Construct a third optimization function corresponding to the total number of overdue segments. Construct a fourth optimization function corresponding to the completion time difference of sub-segments within the same parent segment. ;in, Indicates the segmented processing cycle. Indicates the deadline for completion of each segment. Indicates the total number of segments. Indicates the start time of the segment processing. This indicates the parent segment to which the segment belongs.

[0065] In one embodiment, the initial scheduling unit 300 is specifically used to: determine a preset number of fetal position scheduling time periods based on the fetal position attribute data; calculate the latest start time for processing each ship segment based on the segment processing cycle and segment completion deadline; determine the segment overdue risk priority for each ship segment based on the latest start time for processing each ship segment and the fetal position scheduling time periods; and schedule each ship segment according to the scheduling constraints and the segment overdue risk priority to generate the initial scheduling strategy.

[0066] In one implementation, scheduling each vessel segment according to the segment delinquency risk priority includes: if there are segments of the same level with the same delinquency risk priority, then the segments of the same level are scheduled in a random order.

[0067] In one embodiment, the optimization scheduling unit 500 is specifically used for: determining the non-dominated order of the initial gene sequence based on the calculation result of the scheduling optimization function; calculating the distribution density of the initial gene sequence in the same non-dominated layer according to the non-dominated order, and determining the crowding degree of the initial gene sequence in the same non-dominated layer; selecting the initial gene sequence with a high non-dominated level to obtain an iterative gene sequence, the iterative gene sequence being used to update and replace the initial gene sequence in the next round of iterative genetic operations; and selecting the initial gene sequence with a high crowding degree among the initial gene sequences in the same non-dominated layer to obtain the iterative gene sequence.

[0068] In one embodiment, the optimization scheduling unit 500 is further configured to: for the iterative gene sequence, add weights to the gene loci corresponding to the overdue segments, perform weighted random selection, determine the crossover site, and exchange the gene loci in the two iterative gene sequences that match the crossover site to generate a new iterative gene sequence.

[0069] In one embodiment, the optimization scheduling unit 500 is further configured to: for the iterative gene sequence, add weights to the gene loci corresponding to the overdue segments, perform weighted random selection, determine the mutation sites, and exchange the gene loci in one of the iterative gene sequences that match the mutation sites to generate a new iterative gene sequence.

[0070] The various units described in the above embodiments can be implemented by a computer chip or by a product with a certain function. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0071] For ease of description, the above devices are described separately by function as various units. Of course, in implementing this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0072] Please see Figure 11 This disclosure also provides an electronic device, which includes a memory and a processor. The memory is used to store a computer program, which, when executed by the processor, implements the above-described method for tire position scheduling in ship construction.

[0073] This disclosure also provides a computer-readable storage medium for storing a computer program that, when executed by a processor, implements the above-described method for tire position scheduling in ship construction.

[0074] The processor can be a central processing unit (CPU). It can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or combinations thereof.

[0075] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as the program instructions / modules corresponding to the methods in the embodiments of this disclosure. The processor executes various functional applications and data processing by running the non-transitory software programs, instructions, and modules stored in the memory, thereby implementing the methods in the above-described embodiments.

[0076] The memory may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created by the processor, etc. Furthermore, the memory may include high-speed random access memory and non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include memory remotely located relative to the processor, which can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0077] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk drive (HDD), or solid-state drive (SSD), etc.; the storage medium can also include combinations of the above types of memory.

[0078] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, embodiments of apparatus, devices, and storage media are basically similar to method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0079] The above description is merely an embodiment of this application and is not intended to limit the scope of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

[0080] Although embodiments of the present disclosure have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for scheduling tire positions in shipbuilding, characterized in that, The method includes: Acquire ship section data and tire position attribute data, wherein the ship section data characterizes the expected attributes of each ship section; Based on the ship section data and the tire position attribute data, scheduling constraints and scheduling optimization functions are constructed. The scheduling constraints include spatial constraints, functional constraints, and time constraints. The scheduling optimization function includes total completion time, total overdue days, total overdue sections, and the difference in completion time between sub-sections within the same parent section. Based on the aforementioned scheduling constraints and the segment overdue risk priority of each vessel segment, multiple initial scheduling strategies are generated. The initial scheduling strategy is genetically encoded to obtain an initial gene sequence. Each initial gene sequence corresponds to a fetal position distribution scheme. The gene arrangement position in the initial gene sequence corresponds one-to-one with the segment index. Based on the scheduling optimization function, a non-dominated sorting genetic algorithm is used to perform iterative genetic operations on the initial gene sequence to obtain the Pareto optimal solution set; Genetic decoding is performed on the Pareto optimal solution set to determine the fetal position scheduling strategy.

2. The method according to claim 1, characterized in that, The ship section data includes section length, section width, section processing cycle, section completion deadline, parent section to which the section belongs, and optional construction deck for the section. The tire position attribute data includes tire position length, tire position width, and the construction deck where the tire position is located. The construction of scheduling constraints and scheduling optimization functions includes: The functional constraints are constructed such that the construction building where the target tire position is located belongs to the segment selectable construction building of the target segment, and the target tire position is a candidate assigned to the target segment. The spatial constraints are constructed such that the segment length of the target segment is less than the fetal position length of the target fetal position, and the segment width of the target segment is less than the fetal position width of the target fetal position. The time constraints are constructed such that the scheduling periods for multiple target segments on the same target fetal position do not overlap.

3. The method according to claim 1 or 2, characterized in that, The construction of scheduling constraints and scheduling optimization functions includes: Construct the first optimization function corresponding to the total completion time. ; Construct a second optimization function corresponding to the total number of overdue days. ; Construct a third optimization function corresponding to the total number of overdue segments. ; Construct a fourth optimization function corresponding to the completion time difference of sub-segments within the same parent segment. ; in, Indicates the segmented processing cycle. Indicates the deadline for completion of each segment. Indicates the total number of segments. Indicates the start time of the segment processing. This indicates the parent segment to which the segment belongs.

4. The method according to claim 3, characterized in that, Based on the scheduling constraints and the segment delinquency risk priority of each vessel segment, multiple initial scheduling strategies are generated, including: Based on the fetal position attribute data, a preset number of fetal position scheduling time periods are determined; Calculate the latest start time for processing each ship section based on the section processing cycle and section completion deadline; Based on the latest start time of processing for each vessel section and the aforementioned tire position scheduling time period, the priority of section overdue risk for each vessel section is determined. Based on the scheduling constraints, the initial scheduling strategy is generated by scheduling each vessel segment according to the segment overdue risk priority.

5. The method according to claim 4, characterized in that, The scheduling of each vessel segment is based on the priority of segment delinquency risk, including: If there are segments with the same priority of overdue risk, then the segments with the same priority shall be arranged in a random scheduling order.

6. The method according to claim 1, characterized in that, The step of using a non-dominated sorting genetic algorithm based on the scheduling optimization function to perform iterative genetic operations on the initial gene sequence includes: Based on the calculation results of the scheduling optimization function, the non-dominated order of the initial gene sequence is determined; Based on the non-dominated sorting, the distribution density of the initial gene sequences in the same non-dominated layer is calculated, and the crowding degree of the initial gene sequences in the same non-dominated layer is determined. The initial gene sequence with a high non-dominance level is selected to obtain the iterative gene sequence, which is used to update and replace the initial gene sequence in the next round of iterative genetic operations. For the initial gene sequences in the same non-dominated layer, the initial gene sequences with high crowding are selected to obtain the iterative gene sequences.

7. The method according to claim 6, characterized in that, The step of performing iterative genetic operations on the initial gene sequence using a non-dominated sorting genetic algorithm based on the scheduling optimization function further includes at least one of the following: For the iterative gene sequence, weights are assigned to the gene loci corresponding to the overdue segment, weighted random selection is performed to determine the crossover site, and the gene loci in the two iterative gene sequences that match the crossover site are exchanged to generate a new iterative gene sequence; For the iterative gene sequence, weights are assigned to the gene loci corresponding to the overdue segments, weighted random selection is performed to determine the mutation sites, and gene loci in one of the iterative gene sequences that match the mutation sites are exchanged to generate a new iterative gene sequence.

8. A tire position scheduling device for shipbuilding, characterized in that, The device includes: The data input unit is used to acquire ship section data and tire position attribute data, wherein the ship section data represents the expected attributes of each ship section; The data modeling unit is used to construct scheduling constraints and scheduling optimization functions based on the ship segment data and the tire position attribute data. The scheduling constraints include spatial constraints, functional constraints and time constraints. The scheduling optimization function includes total completion time, total overdue days, total overdue segments and the difference in completion time of sub-segments within the same parent segment. The initial scheduling unit is used to generate multiple initial scheduling strategies based on the scheduling constraints and the segment overdue risk priority of each ship segment. A gene coding unit is used to encode the initial scheduling strategy to obtain an initial gene sequence. Each initial gene sequence corresponds to a fetal position distribution scheme. The gene arrangement position in the initial gene sequence corresponds one-to-one with the segment index. An optimized scheduling unit is used to perform iterative genetic operations on the initial gene sequence based on the scheduling optimization function and a non-dominated sorting genetic algorithm to obtain a Pareto optimal solution set. The gene decoding unit is used to perform gene decoding on the Pareto optimal solution set to determine the fetal position scheduling strategy.

9. An electronic device, characterized in that, The electronic device includes a memory and a processor, the memory being used to store a computer program that, when executed by the processor, implements the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images