Ship block / block hoisting optimization scheduling method and system considering total group site arrangement

Abstract

The invention discloses a ship segment / block hoisting optimization scheduling method and system considering total group site arrangement. The ship segment / block hoisting optimization scheduling method comprises the steps of obtaining information of each segment / block of a ship and information of a total group site; regarding each segment / block as a node, creating a directed graph according to a constraint relationship between a front process and a rear process of the segment / block, and forming a feasible transfer set; according to the pheromones and the heuristic factors, calculating the transition probability of ants among different nodes; for the selected subsections / blocks, the hoisting starting time and the hoisting ending time are determined according to the sequential arrangement or reverse arrangement identification and the available time period of the total assembly site of the subsections / blocks; in the optimization process of the ant colony algorithm, each ant forms a complete scheduling sequence in combination with the transition probability of the ants among different nodes and the hoisting starting and ending time of each segment / block; calculating a target function value corresponding to a scheduling sequence formed by ants by using a time calculation function, and performing pheromone updating on the path of the ant colony; and outputting a scheduling scheme until the algorithm converges.

Description

Technical Field

[0001] The present invention relates to the technical field of shipbuilding and production management, and particularly relates to a method and system for optimizing the hoisting and scheduling of ship blocks / sub-assemblies considering the layout of the pre-assembly site. Background Art

[0002] In the shipbuilding industry, the hull is usually composed of several blocks or ErBlocks. The hoisting sequence, mating cycle, and station allocation of blocks or sub-assemblies affect the overall construction efficiency. When the hoisting and scheduling of ship blocks / sub-assemblies are unreasonable, there will be a large number of mutual waiting or resource conflicts between blocks, and problems such as idleness or excessive waiting of stations and hoisting equipment, resulting in process delays and resource waste. At present, simple manual experience and rules are difficult to adapt to complex working conditions such as simultaneous construction of multiple ships, forward / backward arrangement (the timing of pre-processes or post-processes), etc., resulting in poor plan executability.

[0003] The ant colony algorithm (ACO) is a swarm intelligence optimization algorithm with good global search and robustness, and has been widely used in complex combinatorial optimization methods such as vehicle routing planning and Job-Shop scheduling. However, in the hoisting and scheduling of ship blocks / sub-assemblies in the shipbuilding industry, due to multiple factors such as "pre - post" constraint relationships, forward / backward arrangement timing of blocks, pre-assembly platform allocation, and coupling of mating cycles, there is a lack of more targeted application methods. Summary of the Invention

[0004] Object of the Invention: To overcome the drawbacks of process conflicts, resource idleness, and long overall construction period in the existing hoisting and scheduling of ship blocks / sub-assemblies, the present invention proposes a method and system for optimizing the hoisting and scheduling of ship blocks / sub-assemblies considering the layout of the pre-assembly site. By using the ant colony algorithm and comprehensively considering the sequence of processes of multiple ships and multiple blocks (or sub-assemblies), the availability of station / platform resources, and forward / backward arrangement constraints, etc., it can automatically output a globally optimal or approximately optimal hoisting sequence and station allocation plan.

[0005] Technical Solution: A method for optimizing the hoisting and scheduling of ship blocks / sub-assemblies considering the layout of the pre-assembly site includes the following steps:

[0006] Step 1: Obtain information of each ship block / sub-assembly, including the quantity, name, hoisting time, constraint relationships of pre-processes and post-processes, delay time, and forward or backward arrangement identifier of each block / sub-assembly;

[0007] Step 2: Obtain information of the pre-assembly site, including station number, station type, available time period, and occupied status;

[0008] Step 3: Treat each subsection / total section as a node, create a directed graph according to the constraint relationship between its previous process and subsequent process, and form a feasible transfer set;

[0009] Step 4: Initialize the pheromone matrix and related parameters of the ant colony algorithm;

[0010] Step 5: In each round of iteration, establish several ants. Each ant selects a starting node from the feasible transfer set, and selects the next hoisting node among the subsections / total sections where all previous processes are completed and there is no station conflict. According to the pheromone and heuristic factor, calculate the transfer probability of the ant between different nodes; for the selected subsection / total section, determine its hoisting start and end times according to its forward or reverse arrangement identifier and the available time period of the general assembly site;

[0011] Step 6: In the optimization process of the ant colony algorithm, combine the transfer probability of the ant between different nodes and the hoisting start and end times of each subsection / total section. Finally, each ant forms a complete scheduling sequence;

[0012] Step 7: Use the time calculation function to calculate the objective function value corresponding to the scheduling sequence formed by the ant, and update the pheromone of the ant colony path according to the objective function value;

[0013] Step 8: Repeat Steps 5 to 7 until the algorithm converges or reaches the iteration upper limit, and output the scheduling plan, where the scheduling plan includes: the hoisting sequence of each subsection / total section, the hoisting start and end times, and the station allocation situation of the general assembly site.

[0014] Further, for the selected subsection / total section, determine its hoisting start and end times according to its forward or reverse arrangement identifier and the available time period of the general assembly site. The specific operations include:

[0015] For the subsection / total section with a forward arrangement identifier, calculate the hoisting start time of this subsection / total section based on the end time of the previous process plus the delay time, and determine the end time;

[0016] For the subsection / total section with a reverse arrangement identifier, deduce the hoisting end time of this subsection / total section based on the start time of the subsequent process, and calculate the hoisting start time;

[0017] If there is a conflict between the available time period of the station in the general assembly site and the hoisting start and end times, make adjustments to the currently determined hoisting start and end times, or switch to the next available station.

[0018] Further, in Step 5, according to the pheromone and heuristic factor, calculate the transfer probability of the ant between different nodes according to the following formula:

[0019]

[0020] In the formula, τ ij represents the pheromone intensity, η ij represents the heuristic factor, α represents the pheromone importance coefficient, β represents the heuristic importance coefficient, k represents the candidate or feasible next node, and allowed represents the set of nodes that have not been visited by the current ant and satisfy the constraint conditions.

[0021] Furthermore, in step 7, the objective function value is the total completion time, and the maximum value of all segment / total segment completion times is taken as the total completion time.

[0022] Furthermore, in the process of updating the pheromone for the paths of the ant colony according to the objective function value, it includes:

[0023] For all ant solutions in this round of iteration, first perform pheromone evaporation, which is expressed as: τ ij ←ρ×τ ij , in the formula, τ ij is the pheromone intensity, and ρ represents the pheromone evaporation coefficient;

[0024] Strengthen the pheromone on the edges of the better solution or the optimal solution path, which is expressed as:

[0025]

[0026] In the formula, Q0 represents the pheromone strengthening coefficient, and objectiveValue represents the objective function value corresponding to the current path.

[0027] Perform threshold clipping on the updated pheromone.

[0028] The present invention discloses a ship segment / total segment hoisting optimization scheduling system considering the overall group site layout, including:

[0029] An information acquisition module, used to acquire information of each ship segment / total segment, including the quantity, name, hoisting time, constraint relationship between the pre-process and the post-process, delay time, and the forward or reverse arrangement flag of each segment / total segment; and acquire information of the overall group site, including the work station number, work station type, available time period, and occupied status;

[0030] A feasible transfer set formation module, used to regard each segment / total segment as a node, create a directed graph according to the constraint relationship between its pre-process and post-process, and form a feasible transfer set;

[0031] An optimization module, used to perform optimization iteration based on the information obtained by the information acquisition module and the feasible transfer set output by the feasible transfer set formation module, and obtain a scheduling plan, which includes: the hoisting sequence of each segment / total segment, the hoisting start and end times, and the work station allocation situation of the overall group site.

[0032] Further, in the optimization module, the following steps are executed:

[0033] Step1: Initialize the pheromone matrix and related parameters of the ant colony algorithm;

[0034] Step2: In each iteration, several ants are established. Each ant starts from a starting node selected from the set of feasible transfers, and selects the next liftable node in the segmented / general section where all previous processes are completed and there is no station conflict. According to the pheromone and heuristic factor, calculate the transfer probability of the ant between different nodes; for the selected segmented / general section, determine its lift start and end times according to its forward or reverse arrangement identifier and the available time period of the general assembly site;

[0035] Step3: In the optimization process of the ant colony algorithm, combine the transfer probability of the ant between different nodes and the lift start and end times of each segmented / general section. Finally, each ant forms a complete scheduling sequence;

[0036] Step4: Use the time calculation function to calculate the objective function value corresponding to the scheduling sequence formed by the ant, and update the pheromone of the ant colony path according to the objective function value;

[0037] Step5: Repeat Steps 1 to 4 until the algorithm converges or reaches the iteration upper limit, and output the scheduling plan.

[0038] Further, for the selected segmented / general section, determine its lift start and end times according to its forward or reverse arrangement identifier and the available time period of the general assembly site. The specific operations include:

[0039] For the segmented / general section with a forward arrangement identifier, calculate the lift start time of this segmented / general section based on the end time of the previous process plus the delay time, and determine the end time;

[0040] For the segmented / general section with a reverse arrangement identifier, reverse-deduce the lift end time of this segmented / general section based on the start time of the subsequent process, and deduce the lift start time;

[0041] If there is a conflict between the available time period of the station in the general assembly site and the lift start and end times, adjust the currently determined lift start and end times, or switch to the next available station.

[0042] Further, according to the pheromone and heuristic factor, calculate the transfer probability of the ant between different nodes according to the following formula:

[0043]

[0044] In the formula, τ ijDenotes the pheromone intensity, η ij Denotes the heuristic factor, α denotes the pheromone importance coefficient, β denotes the heuristic importance coefficient, k denotes the candidate or feasible next node, and allowed denotes the set of nodes that have not been visited by the current ant and satisfy the constraint conditions.

[0045] Furthermore, the objective function value is the total completion time, and the maximum value of all sectional / total sectional completion times is taken as the total completion time.

[0046] Beneficial effects: The present invention discloses a method for optimizing the lifting and scheduling of ship sections / total sections considering the general assembly site layout, which relates to the technical field of ship manufacturing and production scheduling. This method first analyzes the data of the pre - post process relationship, in - line / reverse - line process requirements, and station resource information of the sections / total sections, and establishes a corresponding algorithm optimization model. Through the dynamic search of the ant path and pheromone update, it can automatically generate the lifting sequence and station allocation plan on the premise of ensuring process constraints and station feasibility. When detecting a station time conflict, the present invention adjusts the lifting time of the section with a fine step size or switches to other available stations to ensure the smooth connection of the lifting process and reduce the waiting time. Compared with the prior art, the present invention has the following advantages:

[0047] (1) High efficiency: Using the ant colony algorithm, it can quickly find an approximate optimal solution for lifting scheduling in a high - dimensional search space;

[0048] (2) Applicability: It is applicable to complex scenarios with multi - ship parallel construction, multi - process routes, and more reverse - line processes;

[0049] (3) Scalability: It can customize more parameters such as station types and resource constraints according to different production scenarios and be transplanted into the multi - process scheduling of other similar manufacturing industries;

[0050] (4) Compared with the existing manual or simple rule - based scheduling methods, the present invention can significantly shorten the total completion time, improve the station utilization rate, and take into account the requirements of multi - ship parallel production. At the same time, it also supports the secondary scheduling of the general assembly platform and can expand or adjust the algorithm parameters according to the actual situation to achieve a wider range of applications and rapid deployment. Description of the Drawings

[0051] Figure 1 It is the overall structural schematic diagram of a ship section / total section lifting optimization scheduling system proposed by the present invention;

[0052] Figure 2 It is the flow schematic diagram of a ship section / total section lifting optimization scheduling method proposed by the present invention during the section and total section lifting scheduling;

[0053] Figure 3Schematic diagram of the forward and reverse staging calculation processes;

[0054] Figure 4 Schematic diagram of station conflict detection and time fine-tuning. Specific implementation manners

[0055] To make the objectives, technical solutions and advantages of the present invention clearer, the following will further explain a method and system for optimizing the hoisting scheduling of ship sections / sub-assemblies considering the layout of the sub-assembly site in combination with the drawings in the present invention.

[0056] Embodiment 1:

[0057] This embodiment proposes a method for optimizing the hoisting scheduling of ship sections / sub-assemblies considering the layout of the sub-assembly site, which mainly includes the following steps:

[0058] Step 1: By reading the information file containing multiple ships, obtain information such as the quantity, name, hoisting time, constraint relationships (FS / SS / FF / SF) of the pre-operation and post-operation, and delay time of each section / sub-assembly; read the information of each station (Platform) or sub-assembly site, including station number, station type, available time period, occupied status, etc.; set the "Mark" identifier for the forward and reverse sections or sub-assemblies for reverse time calculation during scheduling.

[0059] Step 2: Consider each section / sub-assembly as a node, create a directed graph according to the constraint relationships of its pre-operation and post-operation, and save it using a list or array structure to form a feasible transfer set;

[0060] Step 3: Initialize the pheromone matrix and related parameters of the ant colony algorithm, such as the pheromone matrix τ, the number of iterations NC, the pheromone evaporation coefficient ρ, the initial pheromone concentration P0, etc.

[0061] Step 4: In each round of iteration, establish a number of ants. Each ant selects a starting node from the feasible transfer set, and selects the next hoisting node among the sections / sub-assemblies where all pre-operations are completed and there is no station conflict. Calculate the transfer probability of the ant between different nodes according to the pheromone and the heuristic factor; for the selected section / sub-assembly, calculate the actual hoisting start and end times according to its forward or reverse identifier and the available time period of the sub-assembly site. The specific operations include:

[0062] (1) For the forward (Mark = 0) section / sub-assembly, when calculating the hoisting time, perform time calculation based on the end time of the pre-operation plus the delay time;

[0063] (2) For the reverse (Mark = 1) section / sub-assembly, it is necessary to reverse the hoisting end moment of this section / sub-assembly based on the start time of the post-operation;

[0064] (3) The platform has corresponding types and available time periods. If the current section / block needs to occupy a platform type Tn = 1 or 11, etc., it is necessary to determine whether there is a conflict between the available time period of the platform and the lifting period during scheduling; when a conflict occurs, a slight movement (such as +0.01) will be attempted at the current time or switch to the next available platform until a feasible allocation plan is found or no solution is returned.

[0065] According to the feasible pre-completion or post-start times, combined with the transition probabilities between nodes, construct the "path" of the ant, that is, each ant finally forms a complete scheduling sequence (path); when selecting the next feasible section (or block) at each step, in addition to referring to the pheromone concentration, it is also necessary to check whether the platform conflict, forward or reverse arrangement sequence meets the constraints.

[0066] Step 5: Use a time calculation function (such as Calcu_Time() etc.) to obtain the objective function value (such as the total completion time) corresponding to the scheduling sequence formed by the ants. According to the objective function value, update the pheromone of the paths of the ant colony; when performing pheromone update, adopt a combination of global update and local update. First, volatilize a certain proportion of the pheromone on all paths (that is, multiply by ρ) to avoid premature convergence to a local optimum; then strengthen the pheromone on the transfer edges of the optimal or relatively optimal paths, so as to accelerate the convergence of the algorithm and increase the probability of being adopted by other ants in the next iteration.

[0067] The calculation method of the objective function is: after satisfying all constraints such as pre - post, forward - reverse arrangement and platform conflict, take the maximum value of the completion times of all sections / blocks as the total completion time to measure the quality of the scheduling plan.

[0068] Step 6: Repeat Steps 4 - 5 until the algorithm converges or reaches the iteration limit, and output the optimal or approximately optimal lifting order and platform allocation results, including: the lifting sequence and start / end times of each section / block, the platform allocation situation (the platform numbers and start - end times allocated to specific sections / blocks), the critical path (if any) and the shortest completion time. Through multiple iterations, an approximately optimal lifting scheduling plan can be obtained, improving the shipbuilding efficiency and reducing resource waste.

[0069] In the feasibility detection process of the platform or site in this method, if the platform is occupied by other sections / blocks during the current period, a slight search is carried out on the time axis or switch to the next available platform until a conflict - free interval is found. This method also includes a pre - scheduling step for the overall assembly site allocation of sections / blocks. Based on the ant colony algorithm, available platforms are selected and their start and end times are determined to meet the overall assembly cycle requirements of sections / blocks. The method of this embodiment can effectively improve the shipbuilding efficiency, shorten the total construction period and reduce resource waste.

[0070] Example 2:

[0071] The following is a detailed description of the specific implementation of the present invention in combination with exemplary C# source code. The code is mainly included in Form1.cs and uses several classes (such as ErBlock, PeBlock, Platform, etc.) to store segmented information, platform information, and data structures required for the ant colony algorithm.

[0072] I. Data Structure Design:

[0073] 1. The ErBlock class is used to describe the lifting information of large segments (or sub-segments), including:

[0074] Bn: Segment / sub-segment name or number;

[0075] Nd: Time required for lifting;

[0076] Nt: Time required to reach the lifting position;

[0077] prevTask, prevType, prevDelayTime: Record the index, type (FS / SS / FF / SF, etc.) of the previous process and the delay time respectively;

[0078] FS / SS / FF / SF represent Finish - Start relationship / Start - Start relationship / Finish - Finish relationship / Start - Finish relationship respectively;

[0079] nextTask, nextType, nextDelayTime: Record the index, type, and delay time of the subsequent process respectively;

[0080] Bl: Flag indicating whether it has been lifted;

[0081] Zg: Flag indicating whether it affects the wire drawing illumination;

[0082] te: Loading cycle;

[0083] Bpe: Flag indicating whether general assembly is required;

[0084] Tpe: General assembly cycle;

[0085] Npe: Number of workstations occupied;

[0086] Tn[]: Array of workstation types, used to record the preferred and secondary preferred workstation types;

[0087] Nl: Number of general assembly lifts;

[0088] plfm, Pds, Pde: When assigned to a specific workstation, record the workstation number and the occupied time period.

[0089] 2. PeBlock class (for sectional or sub-sectional information): It is used to describe finer-grained sectional information, especially applicable when large sections need to be split into multiple smaller "sub-sections" before the total group. Its fields are similar to those of ErBlock, but there is an additional Mark (0 = in-sequence, 1 = reverse-sequence) flag, which can be scheduled forward or backward according to process requirements.

[0090] 3. Platform class (for station / platform information):

[0091] plfm: Station number or name;

[0092] Tn: Station type number;

[0093] SBn: List of sections / sub-sections occupying this station;

[0094] Sn: Record the index of the corresponding section / sub-section in the algorithm;

[0095] Pds, Pde: List of start and end times when this station is occupied, for conflict detection and time fine-tuning.

[0096] II. Data Structure Design:

[0097] 1. Read sectional / sub-sectional information (ErBlock / PeBlock):

[0098] Open the specified text file through methods such as ReadErBlock(), and read the sectional information line by line; parse the dependencies of pre- and post-processes, such as "FS / SS / FF / SF + delay", etc.; if some sections (PeBlock) have reverse-sequence characteristics, record them in the Mark = 1 field; count the number of sections that have not been lifted, so as to construct feasible path nodes for the ant colony algorithm.

[0099] 2. Read station / platform information (Platform)

[0100] Store information such as the available time periods, station types, and occupied time periods of multiple stations into the corresponding lists through the ReadPlatform() method; if the station has been occupied by certain processes or equipment before, initialize it in Platform.Pds and Platform.Pde; this information will be used later to determine whether sections / sub-sections can use the station during the corresponding time periods and for station conflict detection.

[0101] 3. Initialize the ant colony algorithm

[0102] Pheromone matrix: Initialize a two-dimensional matrix tao[] of N×N or N×N according to the number of sections / sub-sections N to be scheduled;

[0103] Algorithm parameters: such as the number of ants, the maximum number of iterations, the pheromone evaporation coefficient ρ, the initial pheromone concentration P0, the pheromone intensification coefficient Q0, etc.;

[0104] Data structure: Prepare arrays (such as timerb[], timere[], etc.) to store the start and end times of segments / total segments.

[0105] III. Ant Colony Algorithm Scheduling Process

[0106] The core algorithm process of this embodiment can be composed of a main function (such as mainAnt()) and a series of sub-functions (such as Erection(), Pre_Erection(), PE_Lift(), etc.) to complete the lifting optimization of all segments / total segments. The main steps are as follows:

[0107] 1. Extract the nodes to be scheduled

[0108] Filter the read ErBlock or PeBlock arrays, and only retain the segments / total segments that have not been lifted (or need to be scheduled);

[0109] When the general assembly stage is required, execute an additional scheduling strategy for segments with Bpe = 1 or segments that need to be split twice.

[0110] 2. Path construction

[0111] In each iteration, several "ants" are established, and each ant starts from a feasible "starting node";

[0112] Select the next liftable node among the segments where all previous processes are completed and there is no station conflict. The specific selection logic is jointly determined by the pheromone concentration and the heuristic factor:

[0113]

[0114] Among them, τ ij is the pheromone intensity, and η ij is the heuristic factor, which can be defined based on segment characteristics, process priorities, etc.

[0115] 3. Forward / Backward Scheduling Time Calculation

[0116] If Mark = 0 (forward scheduling), the lifting time t start = max(endOfPrev + delay)

[0117] If Mark = 1 (backward scheduling), it is necessary to reverse the calculation based on the start time of the subsequent process. Let t start = t startOfNext -thisSegmentDuration

[0118] When there is an overlap at the same work station, perform fine time adjustment (+0.01 or other micro step sizes) until the idle period of the work station meets the requirements.

[0119] 4. Objective Function and Path Recording

[0120] After constructing a completely feasible lifting sequence, calculate the "overall completion time" of this sequence (the maximum value among the completion times of all segments / sub-assemblies), and a comprehensive score can also be obtained by combining the work station utilization rate, delay rate, etc.

[0121] Store this solution in a local variable for comparison with the global optimal solution.

[0122] 5. Pheromone Update

[0123] For all ant solutions in this iteration, first perform pheromone evaporation (τ ij ← ρ × τ ij );

[0124] Strengthen the pheromone on the edges (segment i → j) of the better solution (or the optimal solution) path:

[0125]

[0126] To prevent oversaturation or over-dilution, threshold clipping can be performed after the update (such as restricting τ ij not exceeding a certain upper limit).

[0127] 6. Iteration Termination and Result Output

[0128] When the number of iterations exceeds the set threshold, or when the change in the optimal value in several consecutive iterations is less than a certain tolerance, it is considered that the algorithm has converged;

[0129] Write the obtained optimal sequence into output files such as Erection.txt and Platform.txt, and write the key scheduling information (start time, end time, work station number, etc.) into the graphical interface or log file for viewing and analysis.

[0130] IV. Work Station Allocation and Sub-assembly Site Optimization

[0131] For segments / sub-assemblies that need to be sub-assembled (such as Bpe = 1), the present invention provides additional work station allocation strategies, specifically including:

[0132] 1. Platform Feasibility Detection:

[0133] Parse the Platform[] list and check whether the platform type Tn meets the requirements of the segment / sub-assembly;

[0134] If the current platform has been occupied by other segments during the expected time period of this segment, try fine time adjustment or switch to the next platform.

[0135] 2. Secondary Ant Colony Optimization:

[0136] A round of ant colony search can be carried out separately for the overall assembly stage (such as Pre_Erection() or PE_Lift()), taking into account the usage time interval of the Platform and the forward / backward arrangement characteristics of the segments;

[0137] When calculating the objective function, it can focus on "platform utilization rate" or "overall assembly completion time", and continuously optimize the platform scheduling scheme through the pheromone update mechanism.

[0138] V. Example Runs and Effects

[0139] 1. Data Preparation:

[0140] Write the block information of multiple ships (such as hoisting time, relationship between pre- and post-processes) in advance in ErBlock.txt;

[0141] List the available time periods and types of several workstations in Platform.txt;

[0142] 2. Start Algorithm Calculation:

[0143] Run mainAnt(), the system first reads the above files and performs scheduling search according to parameters such as the number of ants and the number of iterations set;

[0144] During the process, several intermediate logs (such as Re.txt) will be generated to record the best completion time of each iteration.

[0145] 3. Output Results:

[0146] After successful convergence, display the hoisting start and end times of each segment / block, the platform number used, and the total completion time in Erection.txt or other output files;

[0147] If secondary optimization is performed on the platform overall assembly process, the segment occupancy details of each platform in different time periods will also be given in Platform.txt.

Claims

1. A method for optimizing the scheduling of ship segment / block hoisting considering the overall assembly site layout, characterized in that: The following steps are involved: Step 1: Obtain the information of each section / block of the ship, including the quantity, name, hoisting time, constraint relationship between the preceding process and the following process, delay time, and sequential or reverse identification of each section / block; Step 2: Obtain the information of the general group site, including the workstation number, workstation type, available time period and occupied status; Step 3: Treat each segment / total segment as a node, create a directed graph based on the constraint relationship between its predecessor and successor processes, and form a feasible transfer set; Step 4: Initialize the pheromone matrix and related parameters of the ant colony algorithm; Step 5: In each round of iteration, a number of ants are established. Each ant selects a starting node from the feasible transfer set, selects the next hoistable node in the segment / total segment where all the previous processes are completed and there is no station conflict, and calculates the transfer probability of the ant between different nodes based on pheromones and heuristic factors; for the selected segment / total segment, determine its hoisting start and end time according to its forward or reverse row mark and the available time period of the total assembly site; Step 6: In the optimization process of the ant colony algorithm, the transfer probability of ants between different nodes and the start and end time of the hoisting of each section / total section are combined. Finally, each ant forms a complete scheduling sequence; Step 7: Use the time calculation function to calculate the objective function value corresponding to the scheduling sequence formed by the ants, and update the pheromone of the ant colony's path according to the objective function value; Step 8: Repeat steps 5 to 7 until the algorithm converges or reaches the iteration limit, and output a scheduling plan, which includes: the hoisting sequence of each segment / segment, the start and end time of hoisting, and the workstation allocation of the overall assembly site.

2. The method for optimizing the ship segment / block hoisting and dispatching considering the overall assembly site layout according to claim 1 is characterized by: For the selected sub-section / block, the start and end time of hoisting is determined according to its forward or reverse identification and the available time period of the assembly site. The specific operations include: For the sections / blocks with sequential markings, the start time and end time of the hoisting of the sections / blocks are calculated based on the end time of the preceding process plus the delay time; For the sections / blocks with reverse markings, the lifting end time of the section / block is calculated backwards based on the start time of the subsequent process, and the lifting start time is calculated; If the available time period of the workstation at the general assembly site conflicts with the start and end time of the lifting, adjustments are made to the currently determined start and end time of the lifting, or switching is made to the next available workstation.

3. The method for optimizing the ship segment / block hoisting and dispatching considering the overall assembly site layout according to claim 1 is characterized by: In step 5, based on pheromones and heuristic factors, the probability of ants transferring between different nodes is calculated according to the following formula: In the formula, τ ij represents the pheromone intensity, η ij represents the heuristic factor, α represents the pheromone importance coefficient, β represents the heuristic importance coefficient, k represents the candidate or feasible next node, and allowed represents the set of nodes that the current ant has not visited and that meet the constraints.

4. The method for optimizing the ship segment / block hoisting and dispatching considering the overall assembly site layout according to claim 1 is characterized by: In step 7, the objective function value is the total completion time, and the maximum value of all segment / total segment completion times is taken as the total completion time.

5. The method for optimizing the ship segment / block hoisting and dispatching considering the overall assembly site layout according to claim 1 is characterized by: The process of updating the pheromone of the ant colony path according to the objective function value includes: For all ant solutions in this round of iteration, pheromone volatilization is first performed, expressed as: τ ij ←ρ×τ ij , where τ ij is the pheromone intensity, ρ represents the pheromone volatility coefficient; The edges on the path of the better solution or the optimal solution are reinforced with pheromone, which can be expressed as: In the formula, Q0 represents the pheromone reinforcement coefficient, and objectiveValue represents the objective function value corresponding to the current path. Perform threshold clipping on the updated pheromone.

6. A ship block / block hoisting optimization scheduling system considering the overall assembly site layout, characterized by: include: The information acquisition module is used to obtain the information of each section / block of the ship, including the number, name, hoisting time, constraint relationship between the preceding process and the following process, delay time, and sequential or reverse identification of each section / block; and obtain the information of the total assembly site, including the workstation number, workstation type, available time period and occupied status; A feasible transfer set forming module is used to regard each segment / total segment as a node, create a directed graph according to the constraint relationship between its predecessor and successor processes, and form a feasible transfer set; The optimization module is used to perform optimization iteration based on the information obtained by the information acquisition module and the feasible transfer set output by the module to obtain a scheduling plan. The scheduling plan includes: the lifting sequence of each section / total section, the start and end time of lifting, and the workstation allocation of the total assembly site.

7. The ship segment / block hoisting optimization scheduling system considering the overall assembly site layout according to claim 6 is characterized by: In the optimization module, perform the following steps: Step 1: Initialize the pheromone matrix and related parameters of the ant colony algorithm; Step 2: In each round of iteration, a number of ants are established. Each ant selects a starting node from the feasible transfer set and selects the next hoistable node in the segment / total segment where all the previous processes are completed and there is no station conflict. The transfer probability of the ant between different nodes is calculated based on pheromones and heuristic factors. For the selected segment / total segment, its hoisting start and end time is determined according to its forward or reverse row mark and the available time period of the total assembly site. Step 3: In the optimization process of the ant colony algorithm, the transfer probability of ants between different nodes and the start and end time of the hoisting of each segment / total segment are combined. Finally, each ant forms a complete scheduling sequence; Step 4: Use the time calculation function to calculate the objective function value corresponding to the scheduling sequence formed by the ants, and update the pheromone of the ant colony path according to the objective function value; Step 5: Repeat Step 1 to Step 4 until the algorithm converges or reaches the iteration limit, and output the scheduling plan.

8. The ship segment / block hoisting optimization scheduling system considering the overall assembly site layout according to claim 7 is characterized by: For the selected sub-section / block, the start and end time of hoisting is determined according to its forward or reverse identification and the available time period of the assembly site. The specific operations include: For the sections / blocks with sequential markings, the start time and end time of the hoisting of the sections / blocks are calculated based on the end time of the preceding process plus the delay time; For the sections / blocks with reverse markings, the lifting end time of the section / block is calculated backwards based on the start time of the subsequent process, and the lifting start time is calculated; If the available time period of the workstation at the general assembly site conflicts with the start and end time of the lifting, adjustments are made to the currently determined start and end time of the lifting, or switching is made to the next available workstation.

9. The ship segment / block hoisting optimization scheduling system considering the overall assembly site layout according to claim 7 is characterized by: According to pheromones and heuristic factors, the probability of ants transferring between different nodes is calculated according to the following formula: In the formula, τ ij represents the pheromone intensity, η ij represents the heuristic factor, α represents the pheromone importance coefficient, β represents the heuristic importance coefficient, k represents the candidate or feasible next node, and allowed represents the set of nodes that the current ant has not visited and that meet the constraints.

10. The ship block / block hoisting optimization scheduling system considering the overall assembly site layout according to claim 7 is characterized by: The objective function value is the total completion time, and the maximum value of all segment / total segment completion times is taken as the total completion time.

Images