Digital intelligent factory management method and system

Abstract

The invention relates to the technical field of shipbuilding production site management, in particular to a digital intelligent factory management method and system, and the method comprises the steps: building a configurable product structure model, a work decomposition structure model and a technological process model for each ship type, and carrying out the real-time sensing and automatic data collection of parameters of a production site; establishing a uniform resource pool, and marking attributes of resources; dynamically evaluating the requirements of different ship types on shared resources in combination with plans and real-time production states; carrying out resource allocation by using an optimization algorithm; the execution state of each process task is monitored in real time, and key production data is automatically collected; and visually displaying each ship type, the construction progress of each section, the critical path, the resource state and the risk early warning information in real time through a billboard, an instrument panel and a report form. According to the invention, through multi-ship-type unified modeling, real-time data perception and intelligent optimization scheduling, collaborative management, cost reduction and benefit increase in the ship building process are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of shipbuilding production site management, and specifically to a digital intelligent factory management method and system. Background Art

[0002] Due to the highly complex structure of its products (such as oil tankers, container ships, special ships, etc.), the normalization of parallel construction of multiple ship types, and the dynamic and variable production process, the traditional management mode of the shipbuilding industry often falls into information silos due to relying on manual experience, paper documents, and decentralized information systems among various departments. This directly leads to problems such as lag in production plan formulation and adjustment, low utilization rate of shared resources (shipyards, large equipment, professional manpower), difficulties in material and supply chain coordination, and low cross-departmental decision-making efficiency. These problems are particularly prominent during the parallel construction of multiple ship types, seriously affecting the construction cycle, cost, and quality.

[0003] Although emerging information technologies such as the Internet of Things, big data, and AI have brought opportunities for transformation to the manufacturing industry, existing digital solutions often struggle to fully adapt to the project-based, multi-variety, long-cycle, and highly discrete manufacturing characteristics unique to the shipbuilding industry; in particular, there is a lack of a unified digital management system that can manage multiple ship projects with different design and process requirements.

[0004] Therefore, the core technical problem faced by the current shipbuilding industry is the lack of a management method and system for the characteristics of parallel construction of multiple ship types, resulting in low overall management decision-making efficiency, extended ship construction time, continuous increase in construction costs, and difficulty in ensuring product quality stability. Summary of the Invention

[0005] The purpose of the present invention is to provide a digital intelligent factory management method and system, which realizes collaborative management and cost reduction and efficiency improvement in the shipbuilding process through unified modeling of multiple ship types, real-time data perception, and intelligent optimization scheduling. For each ship type, it includes establishing configurable product structure models, work breakdown structure models, and process flow models, performing real-time perception and automated data collection of parameters at the production site; establishing a unified resource pool and marking the attributes of resources; dynamically evaluating the demand for shared resources by different ship types in combination with the plan and real-time production status; using optimization algorithms for resource allocation; real-time monitoring the execution status of each process task and automatically collecting key production data; and visually displaying the construction progress, critical path, resource status, and risk warning information of each ship type and each section in real time through dashboards, dashboards, and reports.

[0006] To achieve the above object, the present invention provides the following technical solutions: A digital intelligent factory management method, characterized by including: Based on the integrated data and business modeling with multi-ship type configuration, establish a unified management data model; for each ship type, establish a configurable product structure model, work breakdown structure model, and process flow model, refined to the section, unit, and even to the specific assembly or process level; support multi-project parallel planning and resource budget allocation; the system can quickly switch and load the data and process configurations of different ship types; Automatically collect data on personnel activities, equipment status, material location and flow, and environmental parameters at the production site; Unified management and dynamic scheduling of multi-ship type resources, establish a unified production resource pool, and mark the attributes of resources; combined with the plan and real-time production status, dynamically evaluate the demand for shared resources of different ship types; use optimization algorithms for resource allocation; Refined production process execution and monitoring, real-time monitor the execution status of each process task, and automatically collect key production data; Through the form of kanban, dashboard and report, visually display the construction progress and risk warning information of each ship type and each section in real time.

[0007] Preferably, the integrated data and business modeling based on multi-ship type configuration includes: establishing a unified management data model for integrating data from multiple business fields such as design, process, plan, procurement, material, production, quality, human resources, and equipment; For each type of ship to be built or under construction, establish a set of corresponding configurable product structure models, work breakdown structure models, and process flow models, where the product structure models, work breakdown structure models, and process flow models are refined to the section, unit, and even to the specific assembly or process level; According to the current construction task or query requirement, quickly switch and load the data and configurations of the product structure model, work breakdown structure model, and process flow model corresponding to different ship types.

[0008] Preferably, the data for the automatic data collection includes: the activity status of the operating personnel at the work site, including but not limited to location, on-duty status, and working hours; the status of the equipment, including but not limited to running status, utilization rate, and fault information; the location and flow status of the materials, including but not limited to inventory, in-process product location, and transportation track; and the environmental parameters at the production site, including but not limited to temperature, humidity, and air quality; Transmit the data collected by real-time perception and automatic collection to the unified management data model.

[0009] Preferably, the unified management and dynamic scheduling of multi-ship type resources include: establishing and maintaining a unified production resource pool, which includes but is not limited to various types of production resources such as personnel, equipment, tooling fixtures, and site space, and recording and tracking the detailed attributes, capabilities, available time, and real-time status of each resource; Based on the production plans, process flows of different ship types, and the real-time production status data collected from the production site, dynamically evaluate the real-time demands of multiple parallel shipbuilding projects for the shared resources in the unified production resource pool; Apply an optimization algorithm to dynamically allocate and schedule the shared resources in the unified production resource pool according to the real-time demands, potential conflicts, and real-time availability of the resources, and optimize the utilization rate of the resources.

[0010] Preferably, the refined production process execution and monitoring include: based on the planning and scheduling results, real-time monitor the execution status of each level of process tasks at the production site, and the process tasks cover the construction stages from block manufacturing, unit assembly, and pre-assembly to sea trial; Through automated means and assisted manual entry, real-time collect the key production data related to the execution of the process tasks. The key production data includes but is not limited to process parameters, equipment operation data, material consumption data, and personnel operation data; transmit the monitored execution status and the collected key production data to a unified management data model for processing and display.

[0011] Preferably, it further includes: Utilize the visualization interface of the digital intelligent factory management system to real-time visually display the operation information from the unified management data model and the processing results of each management module through various forms including but not limited to dashboards, gauges, and reports; the operation information is for each ship type and its respective construction blocks; The visual display is used to assist management personnel and on-site personnel in production monitoring, progress tracking, resource coordination, and decision-making analysis.

[0012] A digital intelligent factory management system includes: A multi-ship data modeling unit, based on the integrated data modeling of multi-ship type configurations, establishes a unified data model; for each ship type, establishes configurable product structure models, work breakdown structure models, and process flow models, refined to the block, unit, and even specific assembly or process levels; the system can quickly switch and load the data and process configurations of different ship types; A multi-ship data collection unit, which real-time senses and automatically collects data on personnel activities, equipment status, material positions and flows, and environmental parameters at the production site; Multi-vessel dynamic resource management unit, establish a unified resource pool, mark the attributes of resources; combine plans and real-time production status to dynamically evaluate the needs of different ship types for shared resources; use optimization algorithms to allocate resources; Multi-vessel monitoring unit monitors the execution status of each process task in real time and automatically collects key production data; The visualization unit displays the construction progress and risk warning information of each ship type and section in real time through dashboards, instrument panels and reports.

[0013] Compared with the prior art, the present invention has the following beneficial effects: 1. By building an integrated data model that supports multiple ship configurations, the data islands that were originally scattered in design, process, planning, procurement, production, quality and other links are connected, providing management with a unified, real-time and transparent information view of all parallel ship projects. Instead of relying on scattered reports and empirical judgments, this integrated data platform can be used to conduct a global review and comparative analysis of the progress, resource requirements and cost status of different ship projects, so as to make more scientific strategic planning and cross-departmental collaborative decisions, and improve the confusion and poor coordination of multi-project management caused by poor information.

[0014] 2. By introducing machine learning to predict future demand, dynamically aggregate demand and quantify uncertainty, and conduct conflict risk assessment based on Monte Carlo simulation methods, the present invention enables managers to foresee potential resource bottlenecks and planning risks earlier. Combined with dynamic priority sorting and iterative optimization algorithms, the system can assist managers in formulating more accurate and robust production plans and resource allocation plans, and in the face of real-time changes and emergencies at the production site, it can quickly adjust plans dynamically and rebalance resources, greatly improving the management and response capabilities to complexity and uncertainty.

[0015] 3. Through the real-time visual dashboard, instrument panel and multi-dimensional analysis report provided by the system, the execution status, material consumption, process compliance and quality data of each ship type, each section and even each key process can be accurately grasped. This enables abnormalities to be discovered immediately, problems to be located quickly, and decisions to be made based on real-time data, thereby truly deepening management into every detail of production and improving the effectiveness of process control and the penetration of management. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a digital intelligent factory management method provided by an embodiment of the present invention; Figure 2 A schematic diagram of a unified management and dynamic scheduling process for multiple ship-type resources provided by an embodiment of the present invention; Figure 3Schematic diagram of the structure of a digital intelligent factory management system provided by an embodiment of the present invention. Detailed implementation manners

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0018] Please refer to Figures 1 to 3 , the present invention provides a digital intelligent factory management method and system, and the technical solutions are as follows:

[0019] Embodiment 1: Shipyard A is a shipbuilding yard. To improve its collaborative management efficiency, optimize resource allocation, enhance project transparency, and strengthen the refined management level during the design and construction of various types of ships, Shipyard A introduced the digital intelligent factory management method provided by the present invention. The flowchart of the method is as Figure 1 shown.

[0020] Based on the integrated data modeling with multi-ship type configuration, a unified data model is established; for each ship type, a configurable product structure model, work breakdown structure model, and process flow model are established, refined to the section, unit, and even the specific assembly or process level; the system can quickly switch and load the data and process configurations of different ship types; In the digital intelligent factory management system deployed at Shipyard A, a framework for initializing and constructing a central unified management data model is established. Subsequently, the system initiates the data integration process using a data adapter adapted to the situation of Shipyard A. This process connects to various existing business information systems within the factory, such as computer-aided design (CAD) systems, enterprise resource planning (ERP) systems, manufacturing execution systems (MES), product lifecycle management (PLM) software, human resources (HR) management modules, and equipment asset management systems. Through these connections, the system periodically extracts various types of raw data from these source systems according to a preset information collection cycle, including ship design specifications, 3D model data, process standard libraries, master production plan details, purchase order data, material master data and inventory information, in-process (WIP) tracking data on the production line, quality inspection results, worker skill matrices, and equipment technical parameters and real-time operating status. Then, these raw data enter the ETL (Extract, Transform, Load) processing stage. For various measurement data and timestamp information extracted from each source system, the system performs standard cleaning of formats and units. For key business data, such as the material attributes of parts, parameter settings in process specifications, and the durations of planned tasks, the system performs missing value checks and logical consistency verifications. If key information is found to be missing, filling suggestions and marks for pending supplementation are made; for values outside the normal business logic range (such as unrealistic equipment parameters or man-hours), they are identified and alarmed, and isolated. For core identity codes and business identifiers (such as job numbers in work orders, equipment numbers, drawing numbers, section numbers, etc.), the system performs existence verification and uniqueness checks. The system verifies whether these codes exist in the corresponding master database and processes duplicate unique identifiers to ensure the accuracy of subsequent data associations. After completing the above data cleaning and standardization processes, these normalized data are loaded and stored in the central unified data model.

[0021] Furthermore, when Shipyard A undertakes a new shipbuilding project, project engineers will execute the following model configuration process in the digital intelligent factory management system: First, they will create a new project configuration instance for the current shipbuilding project. Subsequently, using the modeling tools provided by the system that support multi-ship type configuration, they will start building a complete and interrelated set of exclusive digital models for the said project. In this process, engineers first configure the product breakdown structure model (PBS) of the ship type. This method supports decomposing the physical structure of the ship type from top to bottom level by level to levels such as hull sections and structural units according to the unique design of the ship type. During this process, the system attaches the unique attributes of the ship type (such as specific materials, key dimensions, and precision standards) to the nodes of each level of the PBS, and the configuration fineness of the PBS will reach the section, unit, and even smaller assembly component levels. Then, based on the configured ship type PBS, engineers continue to configure the corresponding work breakdown structure model (WBS). By corresponding the WBS tasks with the PBS structure elements, it supports configuring independent WBS templates and decomposition logics that can be refined to the operation level for different ship types. Subsequently, for the key or typical operations defined in the WBS, engineers will configure detailed process flow models. This configuration includes defining the operation steps of the operation, the types of required resources, key process parameters, and quality control points. It supports configuring and managing different versions of process flow models for similar operations of different ship types, and supports dynamic interaction between these process flow models and real-time equipment data and quality feedback. This systematic configuration process is applied to each different type of ship project undertaken by Shipyard A, thereby generating a set of digital model suites supported by this method for each project, which reflect its characteristics and the fineness reaches the section, unit, and even smaller assembly or single operation levels. These model data are all stored in the unified management data model and are clearly associated with the ship type projects.

[0022] Furthermore, when an authorized user of Shipyard A selects a new target ship type project on the user interface of the digital intelligent factory management system to switch their work context through a specific operation, the request from the user interface triggers the system's dedicated context management service. This service first queries the pre-built and maintained ship type configuration metadata index based on the unique identifier of the received target ship type project. This index library enables the system to locate the storage location of the configuration data packet corresponding to this ship type project in the unified management data model. These meta-information also includes the configuration version number, the last update time, etc. After obtaining the metadata of the target ship type configuration package, the system will perform the core data loading operation. This operation involves batch and efficiently retrieving all the core model data and configuration parameters strictly associated with the target ship type project from the unified management data model. The specific loading content includes the complete PBS hierarchical structure of this ship type and the detailed attributes of its various nodes (such as specific materials, design weights, key dimensional tolerances), the complete WBS task network and the logical dependencies between tasks, as well as the detailed step definitions and parameter settings of all relevant process flow models. In addition to the core model, the system will also load a series of other key configuration data closely related to this ship type project. This includes the currently effective Bill of Materials (BOM) version of this ship type project, the indexes and access paths of relevant drawings and technical specifications in the design document library, the specific cost estimation template and rate table for this ship type project, the inspection plan and acceptance criteria set for this ship type in the quality management module, and the key performance indicator (KPI) benchmarks related to this ship type refined from historical projects, etc.

[0023] Furthermore, to ensure the efficiency of this batch data loading process, when designing the unified data management model, the system has adopted optimization strategies at the database level for the characteristics of ship type configuration data, such as data partitioning, pre-connected views, and appropriate denormalization, and ensured that the database can execute efficient query plans. An application layer caching mechanism may also be used to cache the loaded and infrequently changed ship type configuration data to accelerate subsequent repeated accesses. When all the core data and configurations of the target ship type project are successfully loaded into the memory of the application server, the system will update a system-wide current ship type project context identifier. The change of this identifier will be used as the basis for all subsequent business operations. The change of the current ship type project context identifier will be notified to all business function modules that have registered to listen for this change through internal event publishing and direct API callbacks. These modules include user interface rendering services, scheduling engines, resource allocation modules, material requirements planning calculation services, cost accounting modules, quality traceability modules, and production progress monitoring and reporting modules, etc. When the ship type project context identifier changes, it is notified to all business function modules that have subscribed to this status through internal system messages. After receiving the notification, these modules will immediately start their respective internal status adjustment and data refresh processes. Specifically, each business module will use the new ship type IB as a key parameter to re-query and load the data exclusive to this ship type (such as WBS, BOM, cost data, etc.), and apply its specific configuration parameters and business rules (such as quality standards, process routes). The user interface (UI) will be refreshed to only display the information and operation options of the newly activated ship type project. At the same time, when a business module calls other services, the new ship type IB will be passed as a parameter to ensure that the operations of the entire system are based on the current correct ship type context. Through these coordinated adjustments, the system can quickly and accurately switch to the new ship type project operating environment at each functional level after the user switches the ship type.

[0024] This method addresses the complexity of parallel construction of multiple ship types in the shipbuilding industry by constructing a unified data management model to integrate information across the entire domain and break data islands. It establishes refined and configurable product, work, and process models for each ship type and enables rapid switching of ship type configurations, effectively solving problems such as scattered information, difficult coordination of multiple projects, and slow plan response in the traditional management mode. This not only provides a unified and accurate data foundation for complex shipbuilding, significantly improving the efficiency of plan coordination and decision-making levels, but also lays a solid data foundation for real-time perception, intelligent scheduling, and comprehensive traceability in the subsequent production process.

[0025] Full-element real-time perception and data collection, which can perform real-time perception and automated data collection on personnel activities, equipment status, material positions and flows, and environmental parameters at the production site; Specifically, Shipyard A has deployed diverse data collection technologies and devices in key areas and objects at the production site. This includes installing industrial-grade sensors and Internet of Things (IoT) gateways on main channels, operation stations, and large equipment, equipping key materials and work-in-progress with RFIB tags, and equipping on-site workers and management personnel with mobile intelligent terminals with data collection and input functions. For the real-time perception and data collection of personnel activity status, Shipyard A equips employees with work cards with built-in RFIB chips and combines them with RFIB readers deployed in key operation areas to achieve real-time tracking and automatic recording of personnel positions. When employees enter or leave the operation area or move between workstations, the system automatically records their on-duty status. The collection of working hours combines the barcode scanning at the workstation terminal, the start and end operations of tasks associated with work orders on the mobile terminal, and the method of directly associating some automated equipment with operator information to ensure the accurate association of working hour data with specific operation tasks. For new or key production equipment with digital interfaces, the system obtains information such as its operating status, processing parameters, energy consumption data, and fault codes in real time through direct PLC connection and industrial Ethernet interfaces. For traditional equipment, external sensors such as vibration, current, and temperature are installed, and edge computing gateways are used for preliminary data processing and status judgment, and the utilization rate and fault warning information are uploaded. The maintenance status and fault confirmation of some equipment are also supplemented by the maintenance personnel scanning the code and entering the information on the mobile terminal. At key nodes such as material warehousing, in-warehouse relocation, outbound distribution to workstations, and the flow of products between processes, barcode scanning or RFIB automatic identification technology is applied. For the transfer of large sections and components, in-site positioning technology or vehicle GPS is used to track their transportation trajectories. The inventory data is synchronized in real time through the interface with the warehouse management system. Regarding the environmental parameters at the production site, especially in operation areas with special environmental requirements, such as painting workshops, special material welding areas, or enclosed cabin operation areas, Shipyard A has deployed environmental monitoring sensors. These sensors continuously monitor and automatically record key environmental indicators such as temperature, humidity, and air quality (such as VOC concentration, dust concentration). Finally, all the original data of personnel, equipment, materials, and the environment that are real-time perceived and automatically collected through the above means will undergo data filtering and preliminary format unification. Subsequently, these data are transmitted securely and reliably to the central unified management data model constructed in the first step in a standardized data format through the internal wireless network of the factory.

[0026] Through real-time automated collection of all elements such as personnel, equipment, materials, and environment, it effectively addresses the pain points in traditional shipbuilding management, such as information lag, data dispersion, and unclear on-site situations. The data is real-time imported into a unified model and precisely associated with the context of each ship type project, providing transparent and reliable instant data for dynamically grasping the construction progress of multiple ship types in parallel, material flow, and resource status, strongly supporting the precise adjustment of plans, the optimized scheduling of resources, the rapid response to emergencies, and subsequent quality traceability, thus significantly enhancing the controllability of the production process, decision-making efficiency, and overall operation level.

[0027] Unified management and dynamic scheduling of resources for multiple ship types, establishing a unified resource pool and marking the attributes of resources; combining plans and real-time production status to dynamically evaluate the demand for shared resources by different ship types; using optimization algorithms for resource allocation; its flowchart is as Figure 2 shown.

[0028] Specifically, the system establishes and maintains a unified production resource pool within Shipyard A. The resource pool comprehensively manages all shared resources available for shipbuilding in the factory. The main resource list is shown in Table 1. The system creates a detailed digital file for each resource in the resource pool and updates its availability, load, and health status through the aforementioned real-time data collection process.

[0029] Table 1 List and examples of shared resources in the unified production resource pool

[0030] Furthermore, to enhance the predictability and accuracy of demand assessment, the system of Shipyard A integrates and applies a machine learning prediction model based on LightGBM and quantile regression techniques, which predicts the future demand for key shared resources. Its core input features include: historical consumption data of the same type of tasks for the current ship type, design complexity parameters of the current ship type, construction stage, WBS task attributes, and real-time production and resource status data. Through quantile regression, this prediction model outputs the expected demand quantity and predicted values at multiple quantiles, calculates the probability distribution of future resource demand, and effectively quantifies the uncertainty in the prediction. The model is trained based on the historical data of Shipyard A and is retrained and updated regularly. Subsequently, the system dynamically aggregates the immediate parsed demands and predicted future demands of all parallel ship type projects to form a comprehensive demand load curve for shared resources over a period of time in the future. Specifically, the dynamic aggregation process includes: for each shared resource, the system accumulates a representative quantile value of these probabilistic prediction demands within a unified planning time domain (divided by standard time units) for the immediate parsed demands and probabilistic prediction demands of all parallel ship type projects. Secondly, to ensure that subsequent more refined risk assessments can utilize the complete uncertainty information, the system synchronously collects and prepares to transmit the complete distribution characteristics of these original probabilistic demands. In this way, the aggregation process not only generates a benchmark total demand value but also retains the detailed information on the uncertainty that constitutes this demand. This aggregation process synchronously records the source of each demand (ship type IB, WBS task, project priority), and continuously and automatically recalculates and refreshes it based on the adjustment of the production plan, feedback on the real-time progress, and update of the prediction model, forming a dynamic total demand load view of each shared resource that contains uncertainty information.

[0031] Furthermore, for each shared resource incorporated into the unified resource pool, the system matches and compares the future total demand prediction curve segmented by time obtained through dynamic aggregation with the real-time available capacity of this resource within the same planning time domain retrieved from the resource pool module on a time unit by time unit basis. The direct output of this comparison clearly identifies whether there is a potential gap or bottleneck, potential idle state, or a state of basic supply-demand balance for this shared resource in each future time period. For those time periods and resources identified as having potential demand gaps, the system uses the source information of each demand recorded during the dynamic aggregation process to analyze and reveal which WBS tasks of which ship type projects jointly constitute the competing demands for this scarce resource during the conflict period. At the same time, the system further quantifies the key attributes of this conflict. For example, what is the absolute gap quantity of resource demand, how long is the expected conflict duration, and the importance and priority of the key tasks directly affected by this conflict in their respective ship type projects.

[0032] Furthermore, the system not only performs deterministic supply-demand difference calculations, but also directly applies the Monte Carlo simulation method to quantify the probability of resource conflicts actually occurring during the time period and the various potential degrees of demand exceeding capacity based on the probability distribution characteristics of total demand and the available capacity of resources. Furthermore, the system will evaluate the potential impact of this probabilistic conflict on each ship type project directly involved in the conflict once it occurs, and estimate the delay risk of each key milestone by analyzing the position and importance of the resource on its respective WBS critical path. In addition, the dynamic evaluation process also includes trend analysis and predictive warning functions for resource bottlenecks. The system not only handles conflicts that have been clearly identified in the current or short term, but also continuously analyzes the mid- and long-term demand load evolution trend of shared resources, compares the planning capacity of resources, actively identifies those risk resources with continuous high load or even about to turn into systemic bottlenecks, and generates warnings in advance. The warning information will clearly list the predicted bottleneck resources, the ship type projects that mainly constitute the bottleneck demand, the estimated bottleneck period, and the bottleneck risk level evaluated based on the current information. Ultimately, the output of this dynamic assessment process is a continuously updated, structured analysis of the supply and demand balance of shared resources across the plant, usually presented in the form of a comprehensive report or visual dashboard. This result clearly reveals the current and future load status of each shared resource, identified potential conflict points, and predictive bottleneck resource warnings.

[0033] Furthermore, the optimization algorithm first performs a dynamic priority sorting process for multi-ship type tasks. The system aggregates the WBS tasks in all parallel ship type projects that are in a waiting or schedulable state, and calculates a dynamic comprehensive priority score for each task in real time. This score is calculated based on the preset weights, which are configured and defined by production management experts and planning and scheduling managers in the digital intelligent factory management system of Shipyard A. They select a set of evaluation factors that have a key impact on task scheduling based on the factory's operating strategies in different periods and the management focus of multi-ship type projects, and set adjustable relative weight coefficients for these factors. The system then constructs and executes subsequent comprehensive priority score calculations based on these configured factors and their weights. Company A's input factors include: the strategic baseline priority of the ship type project to which the task belongs, the criticality of the task in the WBS of the ship type to which it belongs, the planned urgency of the task, the real-time completeness of the materials required to complete the task, and the current availability forecast of the resources required for the task. These factors are scored according to the score lookup table, and then weighted to form a real-time rolling, cross-ship type unified task scheduling priority queue.

[0034] Furthermore, when the system attempts to allocate a shared resource to a high-priority candidate task, it will initiate a lightweight, fast simulation for a short time window in the future. In this simulation, the system will tentatively allocate this resource and quickly evaluate the potential impact of this allocation behavior on the availability of this resource for other high-priority subsequent tasks that may come from different ship projects in this time window. If the simulation shows that this allocation will result in a lack of resources for more important tasks in the short term or a serious conflict, the system will mark this potential conflict. When the above-mentioned forward-looking simulation pre-detects a resource conflict, or the existing scheduling needs to be adjusted due to an emergency at the production site, the system will generate several possible scheduling adjustment alternatives based on a preset practical adjustment strategy library for common scheduling problems, which is formulated by an expert team. Subsequently, for each alternative, the system will use the quantitative scoring rules formulated by the engineering staff to quickly calculate its expected positive benefits and possible negative effects. This evaluation result will be presented to the production planner in a quantitative manner to assist him in making a quick and relatively optimized decision; for some simple adjustments with obvious evaluation differences and controllable risks, the system can also perform automatic optimization according to the configuration.

[0035] Furthermore, the entire resource allocation and scheduling process is continuously iterative and self-optimizing. The system processes the resource requirements of tasks in batches according to the dynamically generated task priority queue. After each round of resource allocation is completed, or when the real-time status of the production site is significantly updated, the system will recalculate the task priority based on the latest real-time feedback data, use the preset weights, re-examine the resource load and potential conflicts, and repeat the above resource allocation process as needed.

[0036] Through shared resource pools and real-time status updates, combined with machine learning to quantify demand uncertainty, Monte Carlo simulation is used to assess conflict risks and bottlenecks. Its dynamic priority sorting, lightweight deduction, and iterative optimization algorithm for risk-benefit assessment effectively solve the problems of rigid plans, frequent resource conflicts, and delayed decision-making in the parallel construction of multiple ship types. It realizes forward-looking and refined collaborative management of complex production resources, improves resource utilization, planning accuracy, and adaptability, and provides key support for shipyards to reduce costs, increase efficiency, and ensure delivery.

[0037] Refine the execution and monitoring of the production process, monitor the execution status of each process task in real time, and automatically collect key production data.

[0038] Specifically, the optimized production plan and detailed operation instructions are distributed to each relevant production unit and operator in Shipyard A through the digital system. Subsequently, the system monitors the execution status of each level of process tasks covering the entire shipbuilding stage in real time. The update of the task status is mainly captured through the interactive input of on-site operators on the workshop terminal or mobile devices, as well as the signals directly transmitted back by key equipment on some automated production lines. The system will automatically compare the actual status with the planned progress and highlight the deviations. While the tasks are being executed, the system collects the key production data related to the execution of process tasks in real time through a combination of automated collection and assisted manual input. This mainly includes data such as the process parameters actually applied, the operation data of key equipment, the accurate records of material consumption, the working hours and skill compliance of personnel operations, etc. All the monitored task execution status and the collected key production data are automatically marked with context information such as timestamps, ship type project IB, WBS task numbers, and relevant resource numbers. These structured data are transmitted in real time through the internal network of the factory to the data processing core of the central digital intelligent factory management system and finally stored in the unified management data model, accurately associated with the corresponding ship type projects, WBS tasks, and resource entities. These processed real-time information will be used for subsequent production status visualization, progress tracking, quality analysis, cost control, and management decision-making support.

[0039] Through real-time monitoring and automatic collection of key data, it replaces the traditional manual and paper management, realizing the transparency of the production process. The accurate task status, material consumption, and process parameter data are incorporated into the unified model, providing immediate and accurate basis for the dynamic adjustment, quality traceability, and cost control of parallel construction of multiple ship types, and improving the overall management and control level.

[0040] Through the forms of kanban, dashboard, and reports, the construction progress, critical path, resource status, and risk warning information of each ship type and each section are visually displayed in real time.

[0041] Specifically, in key areas such as production workshops and project management offices, Shipyard A has deployed electronic kanbans. These kanbans are refreshed in real time and visually display the task status (such as to-do, in progress, completed) of specific production lines, operation areas or ship type projects, material distribution, equipment operation status, and immediate quality and safety warnings. The kanban information can be flexibly configured according to needs, focusing on the current operation status of a specific ship type or a certain construction section of it, helping the on-site team quickly understand the progress and bottlenecks. Secondly, the system provides customizable dashboards for managers and professional personnel at different levels. The senior management dashboard can macroscopically display the overall progress of all parallel ship type projects, the comparison between cost and budget, the key resource load rate, and the comprehensive risk level; while the dashboard of the ship type project manager will deeply display the detailed construction Gantt chart, key path status, section completion rate, material completeness analysis, and quality problem statistics of the ship type under his responsibility. These dashboards convert complex data into easy-to-understand insights through various visual elements such as charts, KPI indicators, and trend lines, and support drill-down queries. In addition, the system also supports the generation and review of various structured reports. These reports can be generated on demand or automatically pushed according to a preset cycle, and the content covers detailed progress reports, resource consumption analysis, cost accounting details, quality inspection summaries, and production anomaly statistics of each ship type project. The report data can also be filtered and compared according to dimensions such as ship type, section, and time period, providing data support for in-depth analysis, performance evaluation, and management review. All the operation information presented by these kanbans, dashboards, and reports ensures its real-time nature and accuracy, and is strictly organized and presented according to the context of ship type projects and their construction sections.

[0042] Through kanbans, dashboards and reports, the key information such as the progress, resources, and risks of multi-ship type projects is presented in real time and centrally, effectively breaking the dilemma of information silos and untimely grasp of the overall situation in traditional management. It provides a transparent operation view across ship types and sections for managers at all levels and on-site personnel, thus assisting in achieving rapid and accurate production monitoring and progress tracking, promoting efficient resource coordination, supporting scientific decision-making based on real-time data, and enhancing the overall control efficiency and market response ability of the shipyard in the multi-project parallel environment.

[0043] In view of the technical problem of parallel construction of multiple ship types in the shipbuilding industry, the present invention constructs a unified data management model supporting the configuration of multiple ship types, which improves the problems of data dispersion and decision-making lag caused by information islands in traditional management. By collecting and integrating real-time perception data of all elements, and combining machine learning prediction and optimization algorithms to dynamically evaluate, accurately predict and optimize the scheduling of shared resources, the dilemmas of slow plan adjustment and low resource utilization rate are effectively improved. At the same time, the refined production process plan execution tracking, performance monitoring, and multi-level visual kanban and report system significantly improve the transparency of the production process, the effectiveness of process control, and the scientificity and response speed of management decisions. This digital management method provides a unified and efficient collaborative management platform for shipyards, shortening the construction cycle, reducing costs, and improving quality stability.

[0044] Embodiment 2: Shipyard B is mainly engaged in the construction of various ships. In order to improve the overall construction efficiency, Shipyard B introduces a digital intelligent factory management system provided by the present invention, and its structural diagram is as Figure 3 shown.

[0045] For several types of ships built by Shipyard B, the data modeling unit supports pre-defining and storing the model templates of PBS, WBS, and typical process flows of these standard ship types. When receiving a new order for ships of the same series, project engineers can use the functions of this unit to quickly create the initial model configuration of the new ship type project based on the selected standard template. Subsequently, this unit will focus on supporting the efficient and parametric differential configuration of these models generated based on the template to accurately reflect the customer-selected options in specific orders (such as different ballast water treatment systems, specific coating specifications). The system can clearly manage the derivative relationship and difference points between the basic template and each order instance. In the daily operation of Shipyard B, various design software from different suppliers will be used simultaneously (for example, for some ship types, the hull design software of Supplier A is used, and for some, the pipeline design software of Supplier B is used) or classification society specification checking tools with specific data formats. Therefore, when the data modeling unit executes the unified management of the data model construction and data integration process, it will specifically configure and optimize the data adapters and verification rules for these specific heterogeneous data sources. For each type of ship to be built or under construction, a set of corresponding configurable product structure models (PBS), work breakdown structure models (WBS), and process flow models are established, where the product structure model, work breakdown structure model, and process flow model are refined to the section, unit, and even smaller assembly or process levels; according to the current construction task or query requirement, quickly switch and load the data and configuration of the product structure model, work breakdown structure model, and process flow model corresponding to different ship types.

[0046] By using data collection means such as sensors, Internet of Things devices, automatic identification devices, and mobile terminals, the following key elements at the production site of the shipyard are perceived in real time and automatic data collection is carried out: the activity status of personnel, including but not limited to location, on-duty status, and working hours; the status of equipment, including but not limited to operating status, utilization rate, and fault information; the location and flow status of materials, including but not limited to inventory, in-process product location, and transportation track; and the environmental parameters at the production site, including but not limited to temperature, humidity, and air quality.

[0047] A unified production resource pool is established and maintained using the collected resources. The resource pool includes various types of production resources such as but not limited to personnel, equipment, tooling fixtures, and site space. The detailed attributes, capabilities, available time, and real-time status of each resource are recorded and tracked. Based on this unified resource pool, a multi-ship type dynamic resource management unit performs a dynamic demand assessment of shared resources at Shipyard B. This unit conducts a demand assessment based on the production plans of each parallel ship type project, the configured process flow, and the production status data obtained in real time from the multi-ship type data collection unit. It includes immediately analyzing the direct resource requirements of each ship type WBS task, integrating the LightGBM quantile regression model configured and trained according to the data characteristics of Shipyard B to probabilistically predict future demands, and forming a comprehensive demand load view of shared resources containing uncertainty information through a dynamic aggregation process. Subsequently, the system unit further conducts a comparative analysis of the aggregated demands and the real-time available capabilities of resources, identifies potential conflicts, and uses methods such as Monte Carlo simulation to quantitatively evaluate the risk probability. By executing this series of standardized assessment processes, the system unit at Shipyard B ensures that a structured analysis report on the resource supply-demand balance situation is finally output, providing timely, accurate, and risk-fully considered decision-making inputs for subsequent optimized scheduling.

[0048] Furthermore, according to the preset weights jointly configured and defined by production management experts, project commanders, and the heads of technical support departments, the input factors of Company B include: the strategic level and task confidentiality level of the ship type project to which it belongs, the criticality of the task in the WBS of its ship type, the planned urgency of the task, the real-time completeness status of the materials required to complete the task, the current availability prediction of the resources required for the task, the maturity level of the technology involved in the task and the priority guarantee requirements for test verification, the compliance requirements of the key nodes stipulated in the contract, and the special guarantee level requirements for confidentiality conditions and safety production specifications during task execution. These factors are used to obtain the corresponding scores according to the scoring lookup table, and then weighted calculation is performed to form a real-time rolling, cross-ship type unified task scheduling priority queue.

[0049] Based on the planning and scheduling results, the execution status of each level of process tasks is monitored in real time at the production site, covering the construction stages from block manufacturing, unit assembly, general assembly to sea trials; Through automated means and assisted manual entry, key production data related to the execution of the process tasks are collected in real time, and the key production data include but are not limited to process parameters, equipment operation data, material consumption data and personnel operation data; the monitored execution status and collected key production data are transmitted to the digital intelligent factory management system for processing and display.

[0050] By utilizing the visualization interface of the digital intelligent factory management system, the operational information derived from the unified management data model and the processing results of each management module is displayed in real time and visualized through various forms such as dashboards, instrument panels and reports; the operational information is for each ship type and its construction sections; the visualization display is used to assist management personnel and on-site personnel in production monitoring, progress tracking, resource coordination and decision analysis.

[0051] The digital intelligent factory management system of the present invention deployed by Shipyard B provides a systematic solution for the parallel construction of its diversified conventional merchant ships through its highly integrated functional units and unified data platform. The system effectively reduces the problems of information islands, rigid plans, and inefficient resource allocation in traditional management by virtue of its flexible multi-ship type data modeling, differentiated data collection adapted to the specific needs of Shipyard B, and intelligent demand assessment and dynamic resource optimization scheduling capabilities. This system significantly improves the accuracy of Shipyard B's production plans, cross-ship type collaboration efficiency, and response speed to market changes, achieving leaner production process control and better cost-effectiveness, and providing a powerful digital engine for its continued competitiveness in the conventional merchant ship market.

[0052] By introducing a digital intelligent factory management system provided by the present invention, Shipyard B has improved the overall construction management efficiency, accelerated the construction process, and reduced the construction cost. Compared with the original management system of Shipyard B, the management system provided by the present invention has reduced the construction schedule by an average of 4.6% and the construction cost by an average of 3.2%. This shows that the digital intelligent factory management system provided by the present invention can improve the collaborative operation efficiency and refined management capabilities in ship construction.

[0053] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A digital intelligent factory management method, characterized in that, Including: Based on the integrated data and business modeling of multi-ship type configuration, establish a unified management data model; For each ship type, establish configurable product structure models, work breakdown structure models, and process flow models, refined to the section, unit, and even to the specific assembly or process level; support multi-project parallel planning and resource budget allocation; the system can quickly switch and load the data and process configurations of different ship types; Automatically collect data on personnel activities, equipment status, material location and flow, and environmental parameters at the production site; Unified management and dynamic scheduling of multi-ship type resources, establish a unified production resource pool, and mark the attributes of resources; combined with the plan and real-time production status, dynamically evaluate the demand for shared resources by different ship types; use optimization algorithms for resource allocation; Refined production process execution and monitoring, real-time monitor the execution status of each process task, and automatically collect key production data; Through the form of kanban, dashboard and report, visually display the construction progress and risk warning information of each ship type and each section in real time.

2. The digital intelligent factory management method according to claim 1, characterized in that The integrated data and business modeling based on multi-ship type configuration includes: establishing a unified management data model for integrating data from multiple business fields such as design, process, plan, procurement, material, production, quality, human resources, and equipment; For each type of ship to be built or under construction, establish a set of corresponding configurable product structure models, work breakdown structure models, and process flow models respectively, where the product structure models, work breakdown structure models, and process flow models are refined to the section, unit, and even to the specific assembly or process level; According to the current construction tasks or query requirements, quickly switch and load the data and configurations of the product structure models, work breakdown structure models, and process flow models corresponding to different ship types.

3. The digital intelligent factory management method according to claim 1, wherein The data of the automatic data collection includes: the activity status of the operators at the work site, including but not limited to location, on-duty status, and working hours; the status of equipment, including but not limited to operating status, utilization rate, and fault information; the location and flow status of materials, including but not limited to inventory, in-process product location, and transportation trajectory; and the environmental parameters at the production site, including but not limited to temperature, humidity, and air quality; Transmit the data collected by real-time perception and automatic data collection to the unified management data model.

4. A digital intelligent factory management method according to claim 1, characterized in that The unified management and dynamic scheduling of multi-ship type resources includes: establishing and maintaining a unified production resource pool, the resource pool includes but not limited to various types of production resources such as personnel, equipment, tooling fixtures, and site space, where the detailed attributes, capabilities, available time, and real-time status of each resource are recorded and tracked; Based on the production plans, process flows of different ship types and the real-time production status data collected from the production site, dynamically evaluate the real-time demand for shared resources in the unified production resource pool by multiple parallel ship construction projects; Use optimization algorithms to dynamically allocate and schedule the shared resources in the unified production resource pool according to the real-time demand, potential conflicts, and real-time availability of resources, and optimize the utilization rate of resources.

5. A digital intelligent factory management method according to claim 1, characterized in that The refined production process execution and monitoring include: based on the planning and scheduling results, real-time monitoring of the execution status of process tasks at all levels on the production site, where the process tasks cover the construction stages from block manufacturing, unit assembly, and pre-assembly to sea trial; Through automated means and assisted manual input, real-time collection of key production data related to the execution of the process tasks, where the key production data includes but is not limited to process parameters, equipment operation data, material consumption data, and personnel operation data; Transmitting the monitored execution status and the collected key production data to a unified management data model for processing and display.

6. A digital intelligent factory management method according to claim 1, characterized in that It also includes: Using the visualization interface of the digital intelligent factory management system, real-time visually displaying operation information from the unified management data model and the processing results of each management module through various forms including but not limited to dashboards, gauges, and reports; the operation information is for each ship type and its respective construction blocks; The visual display is used to assist management personnel and on-site personnel in production monitoring, progress tracking, resource coordination, and decision-making analysis.

7. A digital intelligent factory management system that executes a digital intelligent factory management method as described in claim 1, characterized in that, It includes: A multi-ship data modeling unit that establishes a unified data model based on integrated data modeling with multi-ship type configurations; For each ship type, establishing configurable product structure models, work breakdown structure models, and process flow models, refined to the block and unit levels, and down to the specific assembly or process level; the system can quickly switch and load data and process configurations of different ship types; A multi-ship data collection unit that real-time senses and automatically collects data on personnel activities, equipment status, material locations and movements, and environmental parameters at the production site; A multi-ship dynamic resource management unit that establishes a unified resource pool and marks the attributes of resources; combines the plan and the real-time production status to dynamically evaluate the requirements of different ship types for shared resources; uses optimization algorithms for resource allocation; A multi-ship monitoring unit that real-time monitors the execution status of each process task and automatically collects key production data; A visualization unit that real-time visually displays the construction progress and risk warning information of each ship type and each block through dashboards, gauges, and reports.

Images