FMEA-based authority collaborative management system and method

Abstract

The invention relates to the technical field of business management, and particularly discloses an FMEA-based authority collaborative management system and method.The FMEA-based authority collaborative management method is characterized in that project products are disassembled according to composition structures of the project products based on an FMEA method to form an object structure tree, and each node in the object structure tree is associated with specific research and development, production and management tasks; a plurality of task structure trees are formed, then an artificial intelligence algorithm based on deep learning is introduced to perform data analysis on the task structure trees, and deep understanding of task structures is realized based on node attributes of task nodes in the task structure trees and association relationships among the task nodes. Therefore, the task structure tree is intelligently allocated to a proper execution area, and production execution of the task and authority collaborative management between the execution areas are performed on this basis. In this way, more accurate task allocation and resource scheduling can be achieved, efficient execution of tasks and optimal configuration of resources are ensured, and flexibility and efficiency of cross-department cooperation are enhanced.

Description

Technical Field

[0001] The present application relates to the field of business management technology, and more specifically, to a FMEA-based authority collaborative management system and method. Background Art

[0002] With the increasing complexity of global business production chains and the intensification of technological competition, the product or process R&D tasks faced by enterprises are becoming increasingly large and complex. In order to ensure the quality and reliability of products or processes, it is essential to introduce a systematic risk management method in the R&D stage. Traditional project management models often focus on the independent operation of a single department or function, which leads to inefficient collaboration between different functional departments and poor information flow, which in turn affects the overall R&D progress and quality control.

[0003] In this regard, the invention patent with publication number CN116993305A proposes a management and analysis method based on FMEA and multi-person authority collaboration. It disassembles project products based on FMEA, builds an object structure tree, and further distributes the R&D tasks of different node objects on the object structure tree to different R&D functional departments within the cross-functional team. Different department heads have viewing and operation permissions to manage the R&D status of their departments. In addition, heads of different departments can grant each other viewing permissions to achieve online business collaboration, thereby improving R&D efficiency.

[0004] The cross-functional team collaboration mechanism provided by the above solution can solve the problem of managers of different business departments managing business in a decentralized manner in the existing technology, and promote efficient cross-departmental collaboration. However, in the process of product development and production, different tasks require the participation of people with different professional backgrounds. In the existing technology, the allocation of tasks mainly relies on pre-set rules and manual judgment, which is difficult to adapt to the rapidly changing project requirements and technology development trends, and is easily affected by factors such as the personal bias and knowledge blind spots of decision makers, which may lead to waste of resources and affect the progress and quality of the entire project.

[0005] Therefore, an optimized FMEA-based authority collaborative management system and method is expected. Summary of the invention

[0006] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides a kind of authority collaborative management system and method based on FMEA, which disassembles the project product according to its component structure based on the FMEA method to form an object structure tree, and each node in the object structure tree is connected with a specific R&D, production and management task to form several task structure trees, and then further introduces an artificial intelligence algorithm based on deep learning to perform data analysis on the task structure tree, based on the node attributes of each task node in the task structure tree and the association relationship between each task node, a deep understanding of the task structure is achieved, so that the task structure tree is intelligently allocated to the appropriate execution area, so as to facilitate the production execution of the task and the authority collaborative management between the execution areas on this basis. In this way, more accurate task allocation and resource scheduling can be achieved, ensuring the efficient execution of tasks and the optimal allocation of resources, and enhancing the flexibility and efficiency of cross-departmental collaboration.

[0007] According to one aspect of the present application, a method for collaborative authority management based on FMEA is provided, which includes: disassembling project products based on FMEA and constructing an object structure tree; disassembling the object structure tree into several task structure trees, and assigning the task structure tree to execution areas; all the execution areas analyze and manage the production of the task structure trees they are responsible for; all the execution areas access each other based on preset authority rules, and then perform business collaboration.

[0008] In the above-mentioned FMEA-based permission collaborative management method, the object structure tree is disassembled into several task structure trees, and the task structure tree is assigned to the execution area, including: extracting a first task structure tree from the several task structure trees; extracting node attributes of each task node from the first task structure tree; semantically embedding coding the node attributes of each task node to obtain a set of task node attribute semantic embedding coding vectors; performing semantic optimization processing based on medium-granularity semantic association constraints on the set of task node attribute semantic embedding coding vectors to obtain a set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors; based on the association relationship between each task node in the first task structure tree, contextually semantically coding the set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors to obtain a first task structure tree contextual semantic coding matrix; based on the first task structure tree contextual semantic coding matrix, determining the allocation result of the first task structure tree.

[0009] According to another aspect of the present application, a FMEA-based authority collaborative management system is provided, which includes: a task structure extraction module, used to extract a first task structure tree from a plurality of task structure trees; a node attribute extraction module, used to extract node attributes of each task node from the first task structure tree; a semantic embedding coding module, used to perform semantic embedding coding on the node attributes of each task node to obtain a set of task node attribute semantic embedding coding vectors; a semantic optimization module, used to perform semantic optimization processing on the set of task node attribute semantic embedding coding vectors based on medium-granularity semantic association constraints to obtain a set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors; a context semantic coding module, used to perform context semantic coding on the set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors based on the association relationship between each task node in the first task structure tree to obtain a first task structure tree context semantic coding matrix; an allocation result determination module, used to determine the allocation result of the first task structure tree based on the first task structure tree context semantic coding matrix.

[0010] Compared with the prior art, the FMEA-based authority collaborative management system and method provided by the present application, based on the FMEA method, disassembles the project product according to its component structure to form an object structure tree, and links each node in the object structure tree with a specific R&D, production and management task to form several task structure trees, and then further introduces an artificial intelligence algorithm based on deep learning to perform data analysis on the task structure tree, based on the node attributes of each task node in the task structure tree and the association relationship between each task node, to achieve a deep understanding of the task structure, so as to intelligently allocate the task structure tree to the appropriate execution area, so as to facilitate the production execution of the task and the authority collaborative management between the execution areas on this basis. In this way, more accurate task allocation and resource scheduling can be achieved, ensuring the efficient execution of tasks and the optimal allocation of resources, and enhancing the flexibility and efficiency of cross-departmental collaboration. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0012] Figure 1 It is a flowchart of the FMEA-based authority collaborative management method according to an embodiment of the present application.

[0013] Figure 2Schematic diagram of data flow of the FMEA-based permission collaborative management method according to an embodiment of the present application.

[0014] Figure 3 It is a flowchart of sub-step S4 of the FMEA-based authority collaborative management method according to an embodiment of the present application.

[0015] Figure 4 It is a flowchart of sub-step S41 of the FMEA-based authority collaborative management method according to an embodiment of the present application.

[0016] Figure 5 It is a flowchart of sub-step S5 of the FMEA-based authority collaborative management method according to an embodiment of the present application.

[0017] Figure 6 It is a block diagram of the FMEA-based authority collaborative management system according to an embodiment of the present application. DETAILED DESCRIPTION

[0018] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0019] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0020] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0021] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0022] It is worth noting that in this application, all actions to obtain data are carried out in compliance with the relevant data protection laws and policies of the country where the data is located, and with the authorization given by the owner of the corresponding device.

[0023] As mentioned in the background technology above, patent CN116993305A proposes a management and analysis method for multi-person authority collaboration based on FMEA, which includes: disassembling project products based on FMEA and constructing an object structure tree; disassembling the object structure tree into several task structure trees, and assigning the task structure tree to execution areas; all the execution areas analyze and manage the production of the task structure trees they are responsible for; all the execution areas access each other based on preset authority rules, and then conduct business collaboration.

[0024] The cross-functional team collaboration solution provided by the above solution can solve the problem of decentralized management of business by managers of different business departments in the prior art and promote efficient cross-departmental collaboration. However, in the process of product development and production, different tasks require the participation of personnel with different professional backgrounds, and in the prior art, the allocation of tasks mainly relies on pre-set rules and manual judgment, which is difficult to adapt to the rapidly changing project requirements and technical development trends, and is easily affected by factors such as personal bias and knowledge blind spots of decision makers, which may lead to waste of resources and thus affect the progress and quality of the entire project. In response to this technical problem, the present application proposes an optimized FMEA-based authority collaborative management method, which, on the basis of the prior art, further introduces an artificial intelligence algorithm based on deep learning to perform data analysis on the task structure tree, based on the node attributes of each task node in the task structure tree and the association relationship between each task node, realizes a deep understanding of the task structure, thereby intelligently allocating the task structure tree to the appropriate execution area, so as to facilitate the production execution of tasks and the authority collaborative management between execution areas on this basis. In this way, more accurate task allocation and resource scheduling can be achieved, ensuring the efficient execution of tasks and the optimal allocation of resources, and enhancing the flexibility and efficiency of cross-departmental collaboration.

[0025] In the above-mentioned FMEA-based authority collaborative management method, the FMEA method must first be used to conduct a comprehensive risk assessment of the project product. FMEA is a preventive quality management tool that can help identify possible failure modes and their potential impacts, so that measures can be taken in the early stages of design and production to reduce risks. By systematically analyzing the function, failure probability, and consequences of failure of each component, it can be determined which parts require special attention or improvement. The result of this step is a detailed risk list that contains all identified risk points and their priority ranking.

[0026] Next, construct an object structure tree based on the results of FMEA. The object structure tree is a hierarchical representation of the entire project product, which decomposes the product or process according to functional modules, subsystems or other logical units. Each node represents a component of the product or a stage of the process, and the connection between nodes reflects the dependency between nodes. The purpose of constructing an object structure tree is to provide a clear framework for subsequent task decomposition and resource allocation. In this process, it is important to maintain the rationality and integrity of the structure tree, ensure that all key elements are covered, and that the hierarchical relationship is correct.

[0027] With the object structure tree as the basis, we will further break it down into several task structure trees. The task structure tree is a detailed description of a specific functional module or subsystem, which divides each large task into a series of specific work items. These work items constitute the complete R&D process, from preliminary design to final test verification, and each task node carries rich attribute information, such as name, type, estimated time, required resources, responsible person, etc. In this way, the originally complex product or process is converted into multiple relatively independent but interrelated task sets, which are easy to manage and track.

[0028] Figure 1 It is a flowchart of the FMEA-based authority collaborative management method according to an embodiment of the present application. Figure 2 FIG. 1 is a data flow diagram of a FMEA-based permission collaborative management method according to an embodiment of the present application. Figure 1 and Figure 2 As shown, the FMEA-based permission collaborative management method includes the following steps: S1, extracting a first task structure tree from the plurality of task structure trees; S2, extracting node attributes of each task node from the first task structure tree; S3, performing semantic embedding coding on the node attributes of each task node to obtain a set of task node attribute semantic embedding coding vectors; S4, performing semantic optimization processing on the set of task node attribute semantic embedding coding vectors based on medium-granularity semantic association constraints to obtain a set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors; S5, based on the association relationship between each task node in the first task structure tree, performing contextual semantic coding on the set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors to obtain a first task structure tree contextual semantic coding matrix; S6, determining the allocation result of the first task structure tree based on the contextual semantic coding matrix of the first task structure tree.

[0029] In the above-mentioned FMEA-based authority collaborative management method, the step S1 extracts the first task structure tree from the several task structure trees. It should be understood that the project product usually has a complex composition structure, and disassembling it according to the composition structure to establish an object structure tree is helpful to more carefully analyze and manage the R&D analysis tasks of each component. Each node in the object structure tree represents the various components of the project product. In order to facilitate subsequent planning and scheduling, it is further converted into a specific task list according to the functional requirements of each node in the object structure tree, and a task structure tree corresponding to each component is created. By performing independent data analysis on each task structure tree, it is helpful to perform more detailed task understanding and allocation management on the R&D analysis tasks of each component of the project product. Based on this, the present application realizes independent analysis and allocation management of the task structure tree by extracting the first task structure tree from the several task structure trees.

[0030] Specifically, when establishing a task structure tree, comprehensive consideration must be given to factors such as project priority, risk assessment, and resource availability. The project schedule and importance determine which task nodes should be prioritized; the results of FMEA (failure mode and effect analysis) or other risk management tools help identify potential risk points and adjust the order of task nodes accordingly; at the same time, the capabilities and limitations of existing resources must be evaluated to ensure that the selected task nodes can be carried out smoothly under current conditions. In addition, it is also very important to analyze whether there are preconditions or post-conditions between the various task nodes, which helps to plan a reasonable task sequence and avoid unnecessary delays and waste of resources.

[0031] In addition to the quantitative analysis above, qualitative factors such as the professional background and technical capabilities of team members need to be considered to ensure that the team's strengths are fully utilized without exceeding their capabilities. In this way, the team's enthusiasm and work efficiency can be improved, while also reducing the chances of problems caused by skill mismatch.

[0032] Next, it is necessary to test the performance of the extracted first task structure tree throughout the project life cycle through simulation, including but not limited to progress, cost, quality and other aspects. This method can not only discover potential problems in advance, but also provide a basis for optimization solutions. The simulation results should reflect the actual situation as detailed as possible so that more scientific and reasonable judgments can be made. If the first task structure tree reveals obvious defects during the simulation process, it is necessary to return to the previous step to re-evaluate other options to ensure that the established task structure tree is more in line with actual needs and expected goals. Through repeated simulation and optimization, the robustness and adaptability of the task structure tree can be ensured. Finally, when the first task structure tree passes the simulation test, it can be used as the basis for project implementation and subsequent allocation work.

[0033] In the above-mentioned FMEA-based authority collaborative management method, the step S2 extracts the node attributes of each task node from the first task structure tree. It should be understood that the task structure tree contains multiple subtasks involved in the product component production and development process, and each subtask corresponds to a specific node attribute, including but not limited to the type, priority, functional requirements, required professional skills, resource requirements, time limit and other key information of the task node, which is the key basis for task allocation and execution. For example, in an automobile manufacturing project, the node attributes of the "engine assembly" task node include: task name: engine assembly; task type: assembly; priority: high; functional requirements: construction and debugging of engine assembly line; required professional skills: mechanical engineering, electronic engineering; resource requirements: assembly line equipment, professional tools; performance indicators: ensure that the maximum output power of the engine reaches the design value and there is no leakage; potential failure mode: improper installation of piston rings leads to leakage; failure consequences: the engine cannot work normally and the vehicle cannot drive. Therefore, the present application extracts the node attributes of each task node from the first task structure tree, which helps to understand the specific content, execution goals and requirements of each task node in detail, thereby providing detailed data support for subsequent task allocation and resource scheduling.

[0034] Node attributes refer to the information set carried by each task node in the task structure tree, which describes the basic characteristics, requirements, constraints, etc. of the task. For example, a task node may have attributes such as name, type (such as design, test), estimated time, required resources, responsible person, predecessor task, successor task, etc. By extracting and analyzing these attributes, we can have a deeper understanding of the specific content of each task and its position and role in the entire project, so as to achieve refined management, optimize resource allocation, risk assessment and control, improve communication efficiency and enhance decision-making quality.

[0035] In order to effectively extract the node attributes of each task node from the first task structure tree, a systematic process must be followed, that is, to define a set of standard attributes applicable to all task nodes. The attribute set should cover the key elements of the task while also taking into account the special needs of the project. For example, in some cases, in addition to conventional attributes, it may be necessary to add indicators for specific areas, such as security requirements, regulatory compliance, etc.

[0036] Next, for each task node, data is collected according to a predefined set of standard attributes. Considering that the collected data is not always completely accurate or complete, it must be strictly verified and confirmed. On the one hand, the consistency and rationality of the data should be checked to avoid logical errors or obvious deviations; on the other hand, for uncertain or ambiguous parts, further investigation and verification should be carried out, and data should be collected again if necessary. Only verified data can be used as the basis for subsequent work to ensure that the decisions made based on these data are scientific and reasonable.

[0037] After ensuring the accuracy and completeness of the data, you can also consider introducing additional dimensions to enrich node attributes. For example, connect the task node with its related documents, code bases, test cases, etc. to facilitate subsequent queries and use; or add external factors such as geographic location and environmental conditions as auxiliary attributes to help better understand and handle tasks. Such expansion not only enhances the depth and breadth of node attributes, but also provides more flexibility and support for subsequent work.

[0038] Finally, considering that the project is a dynamic process, the content and requirements of the tasks may change as the project progresses. Therefore, it is necessary to establish a mechanism to continuously track and update node attributes to ensure that they always reflect the latest situation. Continuous data update and maintenance not only helps to maintain the timeliness of the data, but also promotes efficient management and high-quality delivery of the entire project.

[0039] In the above-mentioned FMEA-based rights collaborative management method, the step S3 performs semantic embedding coding on the node attributes of each task node to obtain a set of semantic embedding coding vectors of task node attributes. In a specific example of the present application, the step S3 includes: using a semantic embedding encoder based on the Bert model to perform semantic embedding coding on the node attributes of each task node to obtain a set of semantic embedding coding vectors of the task node attributes. It should be understood that since the attribute description of each task node usually contains rich text information, in order to further use a computer model to perform efficient data analysis and processing on these text information, it is necessary to convert the node attributes of each task node into a numerical form that can be understood by a computer. Based on this, the present application uses a semantic embedding encoder based on the Bert model to perform semantic embedding coding on the node attributes of each task node respectively, so as to map the node attributes of each task node to a high-dimensional semantic space, and form a set of semantic embedding coding vectors of task node attributes. It should be known to those of ordinary skill in the art that the Bert model uses a bidirectional Transformer encoder architecture to simultaneously consider the context information of each word in the text sequence, thereby capturing a richer and more accurate semantic representation. When processing the node attributes of task nodes, the Bert model can understand and parse the complex semantic relationships in the attribute descriptions, map them to specific locations in the high-dimensional semantic space, and form vectors with semantic meaning.

[0040] In the above-mentioned FMEA-based collaborative authority management method, the step S4 performs semantic optimization processing on the set of task node attribute semantic embedding coding vectors based on medium-granularity semantic association constraints to obtain a set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors. That is, in order to capture the complex semantic relationships between task nodes at a higher level, such as implicit logical connections, contextual dependencies, and potential interactions, the present application further performs context-aware enhancement processing on the semantic embedding coding vectors of each task node attribute based on a local window, so as to utilize the contextual semantic association information within the local window to enhance the expressive power of the semantic embedding coding vectors of each task node attribute, thereby more accurately understanding the correlation between task nodes and providing a more precise basis for subsequent task allocation and collaborative management. Among them, Figure 3 FIG. 4 is a flowchart of sub-step S4 of the FMEA-based authority collaborative management method according to an embodiment of the present application. Figure 3As shown, the step S4 includes the steps of: S41, calculating the grammatical depth of each task node attribute semantic embedding coding vector in the set of the task node attribute semantic embedding coding vectors to obtain the set distribution of the task node attribute grammatical depth implicit representation values; S42, determining the medium-granularity semantic association window of each task node attribute semantic embedding coding vector in the set of the task node attribute semantic embedding coding vector based on the set distribution of the task node attribute grammatical depth implicit representation values; S43, based on the medium-granularity semantic association window of each task node attribute semantic embedding coding vector, performing medium-granularity contextual semantic enhancement on each task node attribute semantic embedding coding vector to obtain the set of the medium-granularity semantically enhanced task node attribute semantic embedding coding vectors.

[0041] Figure 4 FIG. 4 is a flowchart of sub-step S41 of the FMEA-based authority collaborative management method according to an embodiment of the present application. Figure 4 As shown, the step S41 includes the steps of: S411, mapping each task node attribute semantic embedding coding vector in the set of task node attribute semantic embedding coding vectors to the hyperbolic space respectively to obtain a set of hyperbolic space task node attribute semantic embedding coding vectors; S412, calculating the grammatical depth implicit representation value of each hyperbolic space task node attribute semantic embedding coding vector in the set of hyperbolic space task node attribute semantic embedding coding vectors to obtain the set distribution of the task node attribute grammatical depth implicit representation values.

[0042] More specifically, the step S411 is expressed as follows:

[0043]

[0044] in, represents the set of semantic embedding encoding vectors of the task node attributes, , , , They represent the first, second, and third in the set of semantic embedding encoding vectors of the task node attributes. , task node attribute semantic embedding encoding vector, is the hyperbolic space mapping function, and denote the first mapping weight matrix and the second mapping weight matrix respectively, Indicates the The corresponding hyperbolic space task node attribute semantic embedding encoding vector.

[0045] That is, the present application considers that compared with Euclidean space, the exponentially growing volume of hyperbolic space allows for a more natural embedding of tree structures, and is therefore more suitable for representing hierarchical task structures. Therefore, the present application further maps the semantic embedding encoding vectors of each task node attribute to the hyperbolic space, so as to utilize the special properties of hyperbolic geometry to better represent the hierarchical structure of each task node attribute, and reveal the semantic hierarchy and correlation between the attributes of each task node.

[0046] More specifically, the step S412 is expressed as follows:

[0047] in, represents the square of the norm of a vector, represents the logarithmic function with base 2, represents the grammatical depth measurement function, Indicates the The corresponding task node attribute syntax depth implicitly represents the value.

[0048] That is, by calculating the grammatical depth implicit representation value of the semantic embedding encoding vector of each task node attribute after hyperbolic space mapping, we can analyze its depth in the task structure tree, revealing the importance of each task node attribute in the task structure tree and its hierarchical relationship with other components. For example, in the task structure tree, the greater the depth of a node, the more specific subtasks or operation steps it may involve. In this way, it helps to use this hierarchical information to further strengthen the semantic understanding of the attributes of each task node, thereby providing a more accurate basis for the intelligent allocation of tasks.

[0049] Specifically, the step S42 determines the medium-grained semantic association window of each task node attribute semantic embedding coding vector in the set of the task node attribute semantic embedding coding vector based on the set distribution of the task node attribute grammatical depth implicit representation value. It should be understood that by setting a reasonable semantic association window, it can be ensured that when strengthening the semantic features of each task node attribute, neither important contextual information will be omitted nor too much irrelevant information interference will be introduced, which helps to capture the semantic connection between each task node attribute and other medium-sized task node attributes in a specific context, thereby improving the model's ability to understand complex semantic structures while reducing computational complexity. In a specific example of the present application, the step S42 includes: extracting the grammatical depth implicit representation value of the first task node attribute semantic embedding coding vector in the set of the task node attribute semantic embedding coding vector as the starting position of the medium-granularity semantic association window; along the set distribution of the grammatical depth implicit representation value of the task node attribute, finding the grammatical depth implicit representation value of the task node attribute closest to the grammatical depth implicit representation value of the first task node attribute semantic embedding coding vector as the ending position of the medium-granularity semantic association window; based on the starting position of the medium-granularity semantic association window and the ending position of the medium-granularity semantic association window, determining the medium-granularity semantic association window of the first task node attribute semantic embedding coding vector from the set of the task node attribute semantic embedding coding vector, which is expressed by the formula:

[0050] in, Indicates taking the absolute value, Indicates the index corresponding to the minimum value. Indicates the The medium-grained semantic association window, is the The position index of is the same as The syntax depth of implicitly represents the position index of the task node attribute whose value is closest to the syntax depth of implicitly representing the value. The syntax depth of the task node attribute implicitly represents the distribution of the set of values. The task node attribute syntax depth in each position implies the value.

[0051] Specifically, in a specific example of the present application, the step S43 includes: calculating the semantic association scores of each task node attribute semantic embedding coding vector in the medium-granularity semantic association window of the first task node attribute semantic embedding coding vector relative to the first task node attribute semantic embedding coding vector to obtain a set of first task node attribute semantic association scores; normalizing the set of the first task node attribute semantic association scores to obtain a set of normalized first task node attribute semantic association scores; based on the set of normalized first task node attribute semantic association scores, calculating the position-weighted sum between each task node attribute semantic embedding coding vector in the medium-granularity semantic association window of the first task node attribute semantic embedding coding vector to obtain the medium-granularity semantically enhanced task node attribute semantic embedding coding vector corresponding to the first task node attribute semantic embedding coding vector, which is expressed by the formula:

[0052] in, The first in the set of semantic embedding encoding vectors of the task node attributes task node attribute semantic embedding encoding vector, represents the matrix multiplication operation, is the scoring reference vector, is the scoring reference weight matrix, is the rating bias vector, is the natural exponential function, Indicates the The corresponding medium-grained semantic enhancement task node attribute semantic embedding encoding vector.

[0053] That is, based on the semantic association and hierarchical relationship between each task node attribute and other task node attributes in a specific window, the feature representation of the task node attribute is adjusted to better reflect the actual meaning of each task node attribute in the current context, thereby obtaining a set of medium-grained semantically enhanced task node attribute semantic embedding encoding vectors. In this way, the ability to perceive the semantic dependency relationship of task node attributes in the local area of ​​the task structure tree can be effectively improved, thereby enhancing the accuracy of the semantic representation of each task node attribute.

[0054] In the above FMEA-based permission collaborative management method, the step S5 performs context semantic encoding on the set of the medium-granularity semantically enhanced task node attribute semantic embedding encoding vectors based on the association relationship between the task nodes in the first task structure tree to obtain the context semantic encoding matrix of the first task structure tree. Figure 5 FIG. 5 is a flowchart of sub-step S5 of the FMEA-based authority collaborative management method according to an embodiment of the present application. Figure 5As shown, the step S5 includes the steps of: S51, extracting the association relationship between each task node from the first task structure tree to obtain an association relationship matrix; S52, inputting the set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors and the association relationship matrix into a task tree context semantic encoder based on a graph convolutional neural network model to obtain the first task structure tree context semantic coding matrix.

[0055] Specifically, the step S51 extracts the correlation between each task node from the first task structure tree to obtain a correlation matrix. It should be understood that in the task structure tree, there are usually direct or indirect dependencies between each task node. For example, the start of node a1 may depend on the prior completion of node a2, the task of node a3 can be started after the task of node a1 is completed, node a4 can be executed in parallel with node a5 and there is a certain resource sharing relationship between them, etc. These dependencies constitute the logical framework of task execution. Therefore, in order to clearly depict the predecessor, successor or parallel relationship between each task node and other tasks, the present application further extracts the correlation between each task node from the first task structure tree to construct a correlation matrix. In the correlation matrix, each element represents the degree and type of dependence between the corresponding two task nodes. For example, a1 is the predecessor task of a2, then in the correlation matrix A is 1. a2 is the predecessor task of a3, so is 1. a1 and a4 are parallel tasks, and there is a certain resource sharing relationship between them. and In this way, the dependency between task nodes can be intuitively expressed, thus providing a basis for the reasonable arrangement and scheduling of tasks.

[0056] Specifically, in step S52, the set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors and the association relationship matrix are input into the task tree context semantic encoder based on the graph convolutional neural network model to obtain the first task structure tree context semantic coding matrix. That is, in order to further utilize the association relationship between each task node, perform global context semantic association coding on each task node attribute to achieve global hierarchical task tree structure understanding, the present application further uses the graph convolutional neural network model to process the set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors and the association relationship matrix. In the technical solution of the present application, the set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors is used as the feature input of each node in the graph structure, and the association relationship matrix is ​​used as the side information of the graph structure. The graph convolutional neural network model can aggregate the information of neighbor nodes at different levels to update the feature representation of the central node through the information propagation mechanism, thereby effectively capturing the direct and indirect associations between task nodes, and achieving global context perception of the first task structure tree to obtain the first task structure tree context semantic coding matrix. In this way, we can not only learn the dependencies between task nodes, but also understand the global semantics and position of task nodes in the entire task structure, reveal the task association patterns hidden in the deep structure of the task tree, and help understand the overall layout of the task structure, thereby providing a more comprehensive and in-depth information basis for intelligent scheduling of tasks.

[0057] In the above-mentioned FMEA-based permission collaborative management method, the step S6 determines the allocation result of the first task structure tree based on the context semantic coding matrix of the first task structure tree. In a specific example of the present application, the step S6 includes: inputting the context semantic coding matrix of the first task structure tree into a classifier-based intelligent allocation module to obtain the allocation result, and the allocation result is used to represent the type label of the execution area. Specifically, the classifier learns the best matching pattern between tasks and execution areas through a training data set. In actual applications, after the classifier receives the context semantic coding matrix of the first task structure tree, it performs feature learning on the context semantic coding matrix of the first task structure tree to comprehensively consider the semantic features of the attributes of each task node and the global context information of the task structure tree, and at the same time combines the matching relationship learned in the training process to intelligently allocate tasks to the most appropriate execution area, thereby ensuring efficient execution of tasks and rational use of resources.

[0058] In a preferred example of the present application, inputting the first task structure tree context semantic encoding vector into a classifier-based intelligent allocation module to obtain an allocation result includes: First, the distance between each pair of eigenvalues ​​of the context semantic encoding vector of the first task structure tree is calculated, such as the L2 distance, and the square root of the distance is taken to obtain the context semantic distance representation matrix of the first task structure tree, which is expressed as follows:

[0059] in, Represents the first task structure tree context semantic distance representation matrix The element value at position, , Respectively represent the first task structure tree context semantic encoding vector Position and The eigenvalues ​​at the positions, represents the distance metric function, Represents the context semantic encoding vector of the first task structure tree; Secondly, the first task structure tree context semantic self-association matrix of the first task structure tree context semantic encoding vector as a row vector is obtained, which is expressed by the formula:

[0060] in, The first task structure tree context semantic self-association matrix representing the first task structure tree context semantic encoding vector, represents the context semantic encoding vector of the first task structure tree, and is a row vector, represents the transpose of a vector, Represents matrix multiplication; Next, the first task structure tree context semantic encoding vector is matrix-multiplied with the first task structure tree context semantic distance representation matrix to obtain the first task structure tree context semantic primary mapping vector, which is expressed as:

[0061] in, Represents the context semantic level mapping vector of the first task structure tree, Represents the context semantic distance representation matrix of the first task structure tree; Then, the first task structure tree context semantics primary mapping vector is matrix multiplied with the matrix product of the first task structure tree context semantics distance representation matrix and the first task structure tree context semantics self-association matrix to obtain the first task structure tree context semantics multi-level mapping vector, which is expressed as:

[0062] in, Represents the context semantic multi-level mapping vector of the first task structure tree; Then, the first task structure tree context semantic multi-level mapping vector is interpolated with the first task structure tree context semantic association eigenvector composed of the eigenvalues ​​of the first task structure tree context semantic self-association matrix to obtain an optimized first task structure tree context semantic encoding vector, wherein interpolation or zero padding is performed when the eigenvalues ​​are insufficient; Finally, the optimized first task structure tree context semantic encoding vector is input into a classifier-based intelligent allocation module to obtain an allocation result.

[0063] The present application takes into account that the set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors and the association relationship matrix respectively represent the task node attribute enhanced semantic coding features and the task node attribute associated topological features. When performing task tree context semantic coding based on a graph convolutional neural network model, the semantic coding enhancement of the nodes will cause an imbalance in the graph coding topology between the nodes and the edges, thereby resulting in the lack of instance determination of the context semantic coding feature of the context semantic coding vector of the first task structure tree, thereby affecting the accuracy of the allocation result obtained through the classifier-based intelligent allocation module.

[0064] Therefore, the present application performs a secondary target mapping representation based on a multi-level distribution hierarchy on the self-correlation complete similarity instantiation of the context semantic coding vector of the first task structure tree through a linear target mapping representation of the similarity distance representation matrix of the context semantic coding vector of the first task structure tree obtained by expanding the context semantic coding matrix of the first task structure tree, and compensates for the negative impact factor of association mismatch through the association fusion kernel bias to improve the degree of instance determination of the eigenvalue of the context semantic coding vector of the first task structure tree under similarity constraints, that is, the degree of significance of the eigenvalue as an instance for classification and regression determination, and improves the accuracy of the allocation result obtained by the context semantic coding matrix of the first task structure tree through the classifier-based intelligent allocation module.

[0065] After the tasks are assigned, all execution areas begin to conduct in-depth analysis and production management of the task structure tree for which they are responsible. This stage includes developing a detailed implementation plan, allocating necessary resources, monitoring progress, and solving problems encountered. The execution area must not only ensure that the tasks are completed on time and with high quality, but also maintain close communication with other relevant parties and share information and feedback in a timely manner. This close cooperation helps to form an organic whole and jointly drive the project forward.

[0066] Finally, all execution areas can access each other based on preset permission rules, thereby achieving efficient business collaboration. Permission rules define the interaction between different execution areas, including data sharing, decision-making participation, and other aspects. For example, an execution area may have the right to view the work results of another execution area, but must obtain the other party's consent before making any changes; or, on major decisions, all relevant execution areas need to participate in the discussion and reach a consensus. Such a mechanism not only ensures the smooth flow of information, but also maintains the respective responsibility boundaries.

[0067] In summary, the FMEA-based permission collaborative management method based on the embodiment of the present application is explained, which disassembles the project product according to its component structure based on the FMEA method to form an object structure tree, and each node in the object structure tree is associated with a specific R&D, production and management task to form several task structure trees, and then further introduces an artificial intelligence algorithm based on deep learning to perform data analysis on the task structure tree, based on the node attributes of each task node in the task structure tree and the association relationship between each task node, a deep understanding of the task structure is achieved, so that the task structure tree is intelligently allocated to the appropriate execution area, so as to facilitate the production execution of tasks and the collaborative management of permissions between execution areas on this basis. In this way, more accurate task allocation and resource scheduling can be achieved, ensuring the efficient execution of tasks and the optimal allocation of resources, and enhancing the flexibility and efficiency of cross-departmental collaboration.

[0068] Furthermore, a FMEA-based authority collaborative management system is also provided.

[0069] Figure 6 FIG. 1 is a block diagram of a FMEA-based rights collaborative management system according to an embodiment of the present application. Figure 6As shown, according to the FMEA-based permission collaborative management system 100 of the embodiment of the present application, it includes: a task structure extraction module 110, which is used to extract a first task structure tree from a plurality of task structure trees; a node attribute extraction module 120, which is used to extract node attributes of each task node from the first task structure tree; a semantic embedding coding module 130, which is used to perform semantic embedding coding on the node attributes of each task node to obtain a set of task node attribute semantic embedding coding vectors; a semantic optimization module 140, which is used to perform semantic optimization processing on the set of task node attribute semantic embedding coding vectors based on medium-granularity semantic association constraints to obtain a set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors; a context semantic coding module 150, which is used to perform context semantic coding on the set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors based on the association relationship between each task node in the first task structure tree to obtain a first task structure tree context semantic coding matrix; an allocation result determination module 160, which is used to determine the allocation result of the first task structure tree based on the first task structure tree context semantic coding matrix.

[0070] Here, those skilled in the art can understand that the specific operations of each module in the above-mentioned FMEA-based authority collaborative management system have been referred to above. Figures 1 to 5 The description of the FMEA-based authority collaborative management method has been introduced in detail, and therefore, its repeated description will be omitted.

[0071] The basic principle of the present invention is described above in conjunction with specific embodiments. However, it should be pointed out that the advantages, strengths, effects, etc. mentioned in the present invention are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. must be possessed by each embodiment of the present invention. In addition, the specific details of the above embodiments are only for the purpose of illustration and facilitation of understanding, rather than limitation, and the above details do not limit the present invention to being implemented by adopting the above specific details.

[0072] In the above embodiments, the description of each embodiment has its own emphasis. For the parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed system and method can be implemented in other ways. For example, the system embodiment described above is only schematic. For example, the unit division is only a logical function division, and there may be other division methods in actual implementation. The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0073] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference to a figure in a claim should not be considered as limiting the claim to which it relates.

[0074] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units stated in the system claims can also be implemented by one unit through software or hardware.

[0075] Finally, it should be noted that the above description has been given for the purpose of illustration and description. In addition, the above embodiments are only used to illustrate the technical solution of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A collaborative authority management method based on FMEA, comprising: Disassemble project products based on FMEA and build object structure tree; Decomposing the object structure tree into a plurality of task structure trees, and assigning the task structure trees to execution areas; All the execution areas analyze and manage the task structure tree they are responsible for; all the execution areas access each other based on preset authority rules, and then perform business collaboration, which is characterized in that the object structure tree is disassembled into several task structure trees, and the task structure trees are allocated to the execution areas, including: Extracting a first task structure tree from the plurality of task structure trees; Extracting node attributes of each task node from the first task structure tree; Performing semantic embedding coding on the node attributes of each task node to obtain a set of task node attribute semantic embedding coding vectors; Performing semantic optimization processing on the set of task node attribute semantic embedding coding vectors based on medium-granularity semantic association constraints to obtain a set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors; Based on the association relationship between each task node in the first task structure tree, context semantic encoding is performed on the set of attribute semantic embedding encoding vectors of the medium-granularity semantic enhancement task node to obtain a context semantic encoding matrix of the first task structure tree; Based on the context semantic encoding matrix of the first task structure tree, an allocation result of the first task structure tree is determined.

2. The FMEA-based authority collaborative management method according to claim 1 is characterized in that: The node attributes of each task node are semantically embedded and encoded to obtain a set of task node attribute semantic embedded encoding vectors, including: The node attributes of each task node are semantically embedded and encoded using a semantic embedding encoder based on the Bert model to obtain a set of semantic embedding encoding vectors of the task node attributes.

3. The FMEA-based authority collaborative management method according to claim 2 is characterized in that: The set of the task node attribute semantic embedding coding vectors is subjected to semantic optimization processing based on medium-granularity semantic association constraints to obtain a set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors, including: Calculating the grammatical depth of each task node attribute semantic embedding coding vector in the set of task node attribute semantic embedding coding vectors to obtain a set distribution of implicit representation values ​​of the task node attribute grammatical depth; Determine the medium-grained semantic association window of each task node attribute semantic embedding coding vector in the set of task node attribute semantic embedding coding vectors based on the set distribution of the task node attribute grammatical depth implicit representation values; Based on the medium-granularity semantic association window of each task node attribute semantic embedding coding vector, each task node attribute semantic embedding coding vector is respectively subjected to medium-granularity contextual semantic enhancement to obtain a set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors.

4. The FMEA-based authority collaborative management method according to claim 3 is characterized in that: Calculating the grammatical depth of each task node attribute semantic embedding coding vector in the set of task node attribute semantic embedding coding vectors to obtain a set distribution of implicit representation values ​​of the task node attribute grammatical depth, including: Mapping each task node attribute semantic embedding coding vector in the set of task node attribute semantic embedding coding vectors to a hyperbolic space to obtain a set of hyperbolic space task node attribute semantic embedding coding vectors; The grammatical depth implicit representation value of each hyperbolic space task node attribute semantic embedding coding vector in the set of the hyperbolic space task node attribute semantic embedding coding vector is calculated to obtain the set distribution of the grammatical depth implicit representation values ​​of the task node attribute.

5. The FMEA-based authority collaborative management method according to claim 4 is characterized in that: Based on the set distribution of the task node attribute grammatical depth implicit representation values, determining the medium-grained semantic association window of each task node attribute semantic embedding coding vector in the set of the task node attribute semantic embedding coding vectors includes: Extracting the syntax depth implicit representation value of the first task node attribute semantic embedding coding vector from the set of task node attribute semantic embedding coding vectors as the starting position of the medium-granularity semantic association window; Along the set distribution of the task node attribute grammatical depth implicit representation values, find the task node attribute grammatical depth implicit representation value that is closest to the grammatical depth implicit representation value of the first task node attribute semantic embedding coding vector as the termination position of the medium-granularity semantic association window; Based on the starting position of the medium-granularity semantic association window and the ending position of the medium-granularity semantic association window, the medium-granularity semantic association window of the first task node attribute semantic embedding coding vector is determined from the set of task node attribute semantic embedding coding vectors.

6. The FMEA-based authority collaborative management method according to claim 5 is characterized in that: Based on the medium-granularity semantic association window of each task node attribute semantic embedding coding vector, each task node attribute semantic embedding coding vector is respectively subjected to medium-granularity contextual semantic enhancement to obtain a set of medium-granularity semantic enhanced task node attribute semantic embedding coding vectors, including: Calculating the semantic association score of each task node attribute semantic embedding coding vector in the medium-granularity semantic association window of the first task node attribute semantic embedding coding vector relative to the first task node attribute semantic embedding coding vector to obtain a set of first task node attribute semantic association scores; Normalizing the set of the first task node attribute semantic association scores to obtain a set of normalized first task node attribute semantic association scores; Based on the set of normalized first task node attribute semantic association scores, the position-weighted sum of each task node attribute semantic embedding coding vector in the medium-granularity semantic association window of the first task node attribute semantic embedding coding vector is calculated to obtain a medium-granularity semantically enhanced task node attribute semantic embedding coding vector corresponding to the first task node attribute semantic embedding coding vector.

7. The FMEA-based authority collaborative management method according to claim 6 is characterized in that: Based on the association relationship between each task node in the first task structure tree, context semantic encoding is performed on the set of attribute semantic embedding encoding vectors of the medium-granularity semantic enhancement task node to obtain a context semantic encoding matrix of the first task structure tree, including: Extracting the association relationship between each task node from the first task structure tree to obtain an association relationship matrix; The set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors and the association relationship matrix are input into a task tree context semantic encoder based on a graph convolutional neural network model to obtain the first task structure tree context semantic coding matrix.

8. The FMEA-based authority collaborative management method according to claim 7 is characterized in that: Determining the allocation result of the first task structure tree based on the context semantic encoding matrix of the first task structure tree includes: The first task structure tree context semantic encoding matrix is ​​input into a classifier-based intelligent allocation module to obtain the allocation result, and the allocation result is used to represent the type label of the execution area.

9. A FMEA-based rights collaborative management system, which can implement the FMEA-based rights collaborative management method as described in any one of claims 1 to 8, characterized in that: include: A task structure extraction module, used for extracting a first task structure tree from a plurality of task structure trees; A node attribute extraction module, used to extract node attributes of each task node from the first task structure tree; A semantic embedding coding module, used for performing semantic embedding coding on the node attributes of each task node to obtain a set of task node attribute semantic embedding coding vectors; A semantic optimization module, used for performing semantic optimization processing on the set of task node attribute semantic embedding coding vectors based on medium-granularity semantic association constraints to obtain a set of medium-granularity semantically enhanced task node attribute semantic embedding coding vectors; A context semantic encoding module, used for performing context semantic encoding on the set of attribute semantic embedding encoding vectors of the medium-granularity semantically enhanced task nodes based on the association relationship between the various task nodes in the first task structure tree to obtain a context semantic encoding matrix of the first task structure tree; The allocation result determination module is used to determine the allocation result of the first task structure tree based on the context semantic encoding matrix of the first task structure tree.

Images