Method, apparatus, storage medium and product for ma fmea analysis of procedural material

Abstract

This application relates to the field of industrial quality management and risk analysis, and discloses a MaFMEA analysis method, equipment, storage medium, and product for process materials. The method includes: constructing a synthetic analysis graph, which includes material nodes and process nodes, and the material nodes and process nodes are connected by directed lines to form a directed acyclic graph structure; traversing the synthetic analysis graph, extracting material nodes from the synthetic analysis graph, and constructing a hierarchical formulation analysis structure; extracting process nodes from the synthetic analysis graph, and constructing a time-series process analysis structure; mapping the reaction parameters and characteristic indicators defined at each node in the synthetic analysis graph to corresponding nodes in the process analysis structure and formulation analysis structure, respectively, constructing MaFMEA data for process materials, and performing failure mode and effects analysis, thereby providing a more intelligent and rigorous risk management tool for the research and development and production of process materials.

Description

Technical Field

[0001] This application relates to the field of industrial quality management and risk analysis, and in particular to a MaFMEA analysis method, equipment, storage medium and product for process materials. Background Technology

[0002] Failure Mode and Effects Analysis (FMEA) is a risk management tool widely used in industrial fields. It aims to systematically identify potential failure modes and their consequences during product design or manufacturing, and to develop corresponding preventative measures. Currently, most mainstream FMEA analysis tools and software are developed based on the logical architecture of discrete manufacturing industries (such as automotive parts, mechanical assemblies, and electronic assembly). Their core data models are typically based on static Bill of Materials (BOM) structure trees and independent process flow diagrams, assuming that the product is assembled from defined components through physical assembly.

[0003] However, in process industries such as chemicals, pharmaceuticals, food, and new materials, the product formation process differs fundamentally from discrete manufacturing. The research and development and production of process materials are not simply component assembly, but involve complex physical / chemical reaction processes (such as synthesis, pyrolysis, and polymerization). Existing FMEA analysis tools have the following significant technical limitations when applied to process materials:

[0004] First, existing tools cannot effectively characterize the strong coupling relationship between "formulation-reaction-process". They often treat design failures and process failures as two separate modules. In the development of process materials, the final characteristics of the product (such as purity, particle size, and specific capacity) depend not only on the raw material formulation but also directly on the process parameters (such as temperature, pressure, and reaction time). Current tools cannot accurately analyze the physical / chemical reaction relationships between process materials, nor can they analyze the key technical aspects related to the process path selected at the initial design stage. For example, when process temperature fluctuates, existing software cannot automatically correlate and deduce its impact on the chemical reaction mechanism.

[0005] Second, there is a lack of logical correlation between the control points for the characteristics of process materials. These control points are complex, encompassing not only formulation control (such as material selection and component compatibility) but also process control (such as particle size and moisture content), and the two may even influence each other. Current FMEA tools cannot guarantee a strong correlation between these two aspects, leaving them in a fragmented state.

[0006] Third, there is a lack of dynamic evolution logic for failure modes specific to process materials. Failure modes of process materials are highly dynamic and transferable. On the one hand, the sources of failure are complex, including not only formulation failures (such as insufficient component compatibility, impurity introduction, etc.) but also process failures (such as insufficient equipment precision, environmental interference, pilot-scale effects, etc.). On the other hand, failure modes change dynamically with each stage. For example, "impurities" in upstream raw materials may be removed after a specific "purification process," or they may remain due to insufficient process parameters and cause abnormal downstream reactions. Current FMEA tools lack this property flow extrapolation capability based on reaction mechanisms, and cannot simulate the process of impurity transfer or byproduct generation, leading to omissions in the analytical results due to the limitations of human experience.

[0007] Fourth, the construction of failure chains relies on manual labor, which is inefficient and prone to errors. Under existing analysis models, establishing a complete failure network from "process parameter fluctuations → intermediate property changes → final product functional loss" requires engineers to manually connect numerous nodes. Faced with complex chemical synthesis pathways, manual connection is not only inefficient but also fails to encompass all potential causal logic, resulting in FMEA reports that often fail to fully reflect the true risk status of process materials.

[0008] In summary, current FMEA methodologies and tools are insufficient to analyze and generate complete FMEA reports for process-oriented materials. Therefore, there is an urgent need for an intelligent FMEA analysis method that can adapt to the characteristics of process-oriented material R&D, automatically construct analytical structures based on synthetic reaction mechanisms, and achieve cross-level risk correlation. Summary of the Invention

[0009] One objective of this application is to provide a MaFMEA analysis method, apparatus, storage medium, and product for process materials, at least to address the technical problem that existing FMEA analysis tools cannot accurately characterize the strong coupling relationship between formulation and process parameters in process materials, and lack the ability to dynamically predict failures based on reaction mechanisms.

[0010] To achieve the above objectives, some embodiments of this application provide the following aspects:

[0011] In a first aspect, this application provides a MaFMEA analysis method for process materials, the method comprising:

[0012] Construct a synthetic analysis graph, which includes material nodes and process nodes, and the material nodes and process nodes are connected by directed lines to form a directed acyclic graph structure;

[0013] Traverse the synthesis analysis graph, extract the material nodes in the synthesis analysis graph, establish the parent-child inheritance logic between materials based on the pointing relationship of the directed connection in the directed acyclic graph, and construct a hierarchical formula analysis structure.

[0014] Traverse the synthetic analysis graph, extract process nodes from the synthetic analysis graph, and construct a time-series process analysis structure based on the topological timing logic of the process nodes in the directed acyclic graph.

[0015] The reaction parameters and characteristic indicators defined at each node in the synthesis analysis diagram are mapped to the corresponding nodes in the process analysis structure and the formulation analysis structure, respectively, to construct process material MaFMEA data;

[0016] Failure Mode and Effects Analysis was performed based on the MaFMEA data of the process materials.

[0017] Secondly, some embodiments of this application also provide an electronic device, the electronic device comprising: one or more processors; and a memory storing computer program instructions, which, when executed, cause the processor to perform the steps of the method described above.

[0018] Thirdly, some embodiments of this application also provide a computer-readable medium having computer program instructions stored thereon, which can be executed by a processor to implement the method described above.

[0019] Fourthly, some embodiments of this application also provide a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the method described above.

[0020] Compared with related technologies, the solution provided in this application constructs a synthesis analysis graph containing material nodes and process nodes, and automatically generates formulation analysis and process analysis structures using the topological structure of a directed acyclic graph. This ensures that the material properties in formulation analysis and the process parameters in process analysis originate from the same synthesis mechanism model. Furthermore, by combining a visual mind-map-style synthesis analysis graph, synthesis analysis, formulation analysis, and process analysis can be switched with a single click to dynamically display the synthesis path, formulation module, process module, and reaction analysis module of the process material, as well as their interactions. This provides a more intuitive display of the synthesis process of the process material, formulation control points, and synergy with the process, helping researchers better understand the impact of various formulation and process parameters on the final product.

[0021] Secondly, by mapping reaction parameters and characteristic indices from the synthesis analysis diagram to the analytical structure, the unique physical / chemical reaction logic of process materials can be simulated. Based on topological temporal logic, this method can track the dynamic evolution of material properties in the reaction chain, such as the introduction, transfer, or removal of impurities. This allows for the identification of potential failure modes that dynamically change with process stages, which are undetectable by traditional static structures, such as byproduct residues or pilot-scale effects, significantly improving the comprehensiveness and accuracy of risk identification.

[0022] Finally, leveraging the connectivity of directed acyclic graphs, automated pathfinding and correlation are achieved from "fluctuations in process elements" to "functional failure of the final product." Compared to existing technologies that rely on engineers to perform numerous tedious and error-prone manual node connections, the embodiments of this disclosure can automatically establish cross-level failure causal chains. This not only significantly reduces the manpower cost of FMEA analysis but also prevents the omission of critical failure paths through system-level logical constraints, providing a more intelligent and rigorous risk management tool for the research and development and production of process materials. Attached Figure Description

[0023] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0024] Figure 1 A flowchart of a MaFMEA analysis method for process materials is provided as an exemplary embodiment of this disclosure;

[0025] Figure 2 A schematic diagram of synthesis analysis in a MaFMEA analysis method for process materials provided as an exemplary embodiment of this disclosure;

[0026] Figure 3 A schematic diagram of the formulation analysis structure in a MaFMEA analysis method for process materials provided as an exemplary embodiment of this disclosure;

[0027] Figure 4 A schematic diagram of the process analysis structure in a MaFMEA analysis method for flow materials provided as an exemplary embodiment of this disclosure;

[0028] Figure 5 An exemplary structural diagram of the electronic device provided for some embodiments of this application. Detailed Implementation

[0029] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0030] Figure 1 A schematic flowchart of a MaFMEA analysis method for process materials provided as an exemplary embodiment of this disclosure, the method comprising:

[0031] S101. Construct a synthetic analysis graph, which includes material nodes and process nodes, and the material nodes and process nodes are connected by directed lines to form a directed acyclic graph structure.

[0032] Specifically, such as Figure 2 First, a synthesis analysis diagram is constructed. Users (such as materials R&D engineers) can define the R&D process by dragging and dropping components on a visual interface. Each component created by the user is called a "node," mainly divided into two categories: one is "material nodes" representing physical substances (such as raw materials, intermediates, final products, catalysts, etc.), and the other is "process nodes" representing physical / chemical reaction operations (such as calcination, pickling, coating, etc.). Different nodes can be identified by different color blocks. Users connect these nodes using directed lines according to the actual reaction sequence. This visual flowchart can be parsed into a directed acyclic graph (DAG) data structure, where nodes represent vertices and lines represent directed edges, ensuring the temporal irreversibility and logical rigor of the reaction process.

[0033] S102. Traverse the synthesis analysis graph, extract the material nodes in the synthesis analysis graph, establish the parent-child inheritance logic between materials based on the pointing relationship of the directed lines in the directed acyclic graph, and construct a hierarchical formula analysis structure.

[0034] Specifically, such as Figure 3 The first path of the above synthesis analysis diagram is traversed. During this process, nodes marked with the attribute "material" are identified and extracted, while "process" nodes in the path are automatically filtered out. The parent-child inheritance logic between materials is derived based on the pointing relationships of the directed lines in the DAG diagram. Specifically, if node A points to node B (possibly through process nodes), node A is determined to be a component or precursor (parent) of node B, thus automatically constructing a hierarchical formula analysis structure tree. This process transforms a linear, mixed-action reaction flow into a structured, clean bill of materials.

[0035] S103. Traverse the synthetic analysis graph, extract the process nodes in the synthetic analysis graph, and construct a time-series process analysis structure based on the topological timing logic of the process nodes in the directed acyclic graph.

[0036] Specifically, such as Figure 4 The composite analysis diagram is traversed again to extract nodes marked with the attribute "process". The preset process number (e.g., ...) is read from each process node. Figure 2 The algorithm uses OP10 and OP20 (or a topology sorting algorithm based on a DAG) to determine the order of process nodes. For parallel branches in the graph (e.g., different intermediates are processed separately), it can integrate them based on topology convergence logic to finally construct a time-series process analysis structure tree. This structure tree clearly shows the complete process route from raw material input to product output.

[0037] S104. Map the reaction parameters and characteristic indicators defined at each node in the synthesis analysis diagram to the corresponding nodes in the process analysis structure and the formulation analysis structure, respectively, to construct process material MaFMEA data.

[0038] Specifically, the elements in the constraint nodes of the composite analysis diagram are mapped to the corresponding 5M1E elements (Man, Machine, Material, Method, Environment, Measurement) under the process analysis structure. The specific constraints under the elements in the constraint nodes (such as reaction temperature, time, acid concentration, etc.) are automatically mapped to the process characteristics under the elements in the process analysis structure. Simultaneously, the characteristic indicators defined on the material nodes (such as purity, particle size, porosity) are automatically mapped to the corresponding "product characteristics" in the formulation analysis structure. Through this mapping, the initialization of the basic data for FMEA analysis is completed, ensuring that the design intent (reaction mechanism) is seamlessly transmitted to the manufacturing end (process control).

[0039] S105. Perform failure mode and effects analysis based on the MaFMEA data of the process material.

[0040] Specifically, based on the structured data constructed above, risk simulations can be performed. For example, in response to parameter fluctuations input by users in the process analysis structure (such as mechanical factors: excessively high furnace temperature), the analysis structure can derive corresponding failure modes (such as "decreased porosity of intermediates"), and further deduce functional failures of the final product (such as "battery specific capacity not meeting standards"), thus forming a complete failure chain and assisting R&D personnel in formulating corresponding preventive measures and optimization plans.

[0041] In the above embodiments, by constructing a synthetic analysis graph containing material nodes and process nodes, and utilizing the topological structure of a directed acyclic graph (DAG) to automatically generate both the formulation analysis structure and the process analysis structure through dual-track parsing, the limitations of traditional FMEA tools that artificially separate the design formulation from the manufacturing process are fundamentally overcome. Furthermore, this generation mechanism ensures that the material properties in the formulation analysis and the process parameters in the process analysis originate from the same synthetic mechanism model, achieving deep coupling and source management of formulation design and process data, effectively avoiding information silos and logical conflicts caused by manually maintaining two separate sets of data. Simultaneously, the DAG provides a data foundation for deriving the complete causal chain of "process element fluctuation → intermediate property change → final product functional failure." Compared to existing technologies that rely on engineers to perform a large amount of tedious and error-prone manual node connections, this method significantly reduces the labor costs of FMEA analysis and prevents the omission of critical failure paths through system-level logical constraints, providing a more intelligent and rigorous risk management tool for the research and development and production of process materials.

[0042] In one embodiment, the step of mapping the reaction parameters and characteristic indices defined at each node of the synthesis analysis diagram to the corresponding nodes in the process analysis structure and the formulation analysis structure further includes:

[0043] Configure material type labels for the material nodes, and map the material nodes to the corresponding classification levels in the recipe analysis structure based on the material type labels;

[0044] Configure element attribution attributes for the reaction constraints of the process nodes, and map the reaction constraints to the corresponding production element categories in the process analysis structure based on the element attribution attributes.

[0045] Specifically, the reaction parameters and characteristic indicators defined at each node in the synthesis analysis diagram are mapped to corresponding nodes in the process analysis structure and formulation analysis structure, further including refined classification mapping based on node attributes. For the mapping of material nodes, it supports configuring material type labels for each material node in the synthesis analysis diagram. These labels are used to distinguish the role of the node in the chemical reaction, such as "main reactant," "catalyst," "main product," and "byproduct."

[0046] Labels such as "solvent" or "additive" are used. When constructing the formulation analysis structure, these labels are parsed, and instead of simply listing all materials at the same level, they are mapped to the corresponding classification levels in the formulation tree based on the labels. It is possible to select a process that outputs only one "main product," excluding byproducts from the structure tree analysis. For example, the manganese node labeled "catalyst" will automatically be assigned to the "auxiliary materials" branch, while the node labeled "carbon source" will be assigned to the "main components" branch, thus forming a clearly structured and hierarchical formulation system.

[0047] Simultaneously, for the mapping of process nodes, element attribution attributes are configured for the reaction constraints (such as temperature, pressure, time, etc.) under each process node. This attribute identifies which category of production element ("Man, Machine, Material, Method, Environment, Measurement") the constraint belongs to in production management. For example, "calcination temperature" can be identified as the precision requirement of the "Machine (Equipment)" element, and "material feeding sequence" can be identified as the operational specification of the "Method (Process Method)" element. When constructing the process analysis structure, based on these element attribution attributes, each reaction constraint is automatically distributed and attached to the specific production element category under the corresponding process. This mapping method ensures that the root cause of failure can be quickly located during subsequent PFMEA (Process Failure Mode and Effects Analysis), such as distinguishing between temperature deviations caused by insufficient equipment capacity and material feeding errors caused by improper human operation.

[0048] In this embodiment, by introducing material type tags, the solution achieves structured and refined management of formulation data. Unlike traditional methods that simply list all materials, this embodiment can automatically distinguish between "main reactants" and auxiliary materials such as "catalysts" and "additives," and map them to different levels in the formulation analysis structure. This classification mechanism not only makes the generated formulation tree more hierarchical and logically rigorous, but also enables R&D personnel to conduct more targeted analyses of specific risk points such as catalyst failure and additive compatibility, avoiding risk omissions due to material confusion. Furthermore, by introducing element attribution attributes, the standardization of process control and the accuracy of failure tracing are improved. The system can automatically classify complex reaction constraints into specific production element categories such as "people, machines, materials, methods, environment, and measurement." This mapping method breaks the limitation of isolated parameters in traditional FMEA, ensuring that each control indicator has a clear management attribution. When process failure occurs, analysts can quickly locate the root cause of the problem (e.g., whether it is insufficient "equipment" precision or "personnel" operational error), thereby greatly improving the efficiency of analysis and the implementation of improvement measures.

[0049] In one embodiment, the step of traversing the synthesis analysis graph, extracting material nodes from the synthesis analysis graph, establishing parent-child inheritance logic between materials based on the pointing relationships of directed connections in the directed acyclic graph, and constructing a hierarchical formula analysis structure further includes:

[0050] Define risk attributes for the material node, including impurity content, by-product risk, or physical property parameters;

[0051] When establishing the parent-child inheritance logic between materials, the risk attributes of upstream material nodes are inherited to downstream material nodes according to the topological relationship of directed connections, forming an attribute transfer flow.

[0052] Specifically, firstly, a set of risk attributes can be defined for each material node in the synthesis analysis graph. These attributes are not simple text labels, but rather a quantifiable set of parameters, including, but not limited to, the impurity content in the raw materials (such as iron ion concentration), the risk level of potential byproducts generated during the reaction, and key physical property parameters (such as moisture content and particle size distribution). Based on this, when establishing the parent-child inheritance logic between materials according to the pointing relationships of directed lines in the directed acyclic graph, the attribute genetic algorithm is activated simultaneously. According to the topological relationship, the risk attributes carried by upstream material nodes are "inherited by default" to downstream material nodes. For example, if the upstream "carbon source" node is marked as containing a specific impurity, the system will automatically pass on this impurity attribute when generating the downstream "intermediate" node, thus forming a coherent attribute transfer flow that simulates material flow throughout the entire formulation analysis structure. This mechanism ensures that the initial risk of raw materials can be tracked throughout the synthesis path, rather than being lost during hierarchical transformations.

[0053] In the above embodiments, by defining risk attributes for material nodes that include quantitative parameters such as impurity content and by-product risk, and automatically inheriting these attributes based on the topological relationships of a directed acyclic graph, this scheme achieves dynamic simulation of material flow in process materials. Unlike traditional FMEA tools that can only construct static structures, this method can form a coherent attribute transfer in the formulation analysis structure. This mechanism enables the system to simulate how trace impurities or physical properties in raw materials are transmitted downstream along the reaction path, thereby accurately characterizing the objective law that "upstream component fluctuations directly affect the properties of downstream products" in process materials. Furthermore, this attribute inheritance mechanism greatly improves the comprehensiveness and traceability of risk analysis. In complex chemical synthesis pathways, the initial risks of raw materials (such as excessive iron ions due to batch differences) are often easily overlooked in the hierarchical transformation of multi-stage reactions. This embodiment, through automated attribute inheritance, ensures that every key risk attribute upstream can be tracked and transmitted to intermediate and final product nodes throughout the entire process. This not only effectively avoids the risk of omission due to human error in analysis, but also provides accurate data support for subsequent analysis of "cumulative effect" or "amplification effect", significantly improving the quality and practical value of the final report.

[0054] In one embodiment, the step of inheriting the risk attributes of upstream material nodes to downstream material nodes by default and forming an attribute transfer flow based on the topological relationship of directed connections when establishing the parent-child inheritance logic between materials further includes:

[0055] When the attribute transfer flow passes through the process node, the process capability definition and corresponding process parameters of the process node are read.

[0056] If the process parameters meet the effective threshold in the process capability definition, the corresponding risk attribute is reduced when generating downstream material nodes;

[0057] If the process parameters do not meet the effective threshold, the process capability is determined to be in failure, and the corresponding risk attribute is maintained when generating downstream material nodes, and a potential failure mode marker is generated in the formula analysis structure.

[0058] Specifically, when an attribute transfer flow is detected to pass through a process node in the synthetic analysis graph (e.g., Figure 2When the process reaches the "OP20 Pickling" or "Caking" node, a verification process is triggered. First, the process capability definition for that node and the actual process parameters configured by the user on that node (such as acid concentration and heating temperature) are read. Then, the user-input process parameters are compared with the effective thresholds in the process capability definition. If the process parameters meet the effective thresholds (e.g., the user-set acid concentration is higher than the minimum concentration required to remove metal ions), the process capability is deemed effective, meaning the risk transmission has been successfully reduced. At this point, when generating downstream material nodes (such as "Porous Carbon"), the risk level of the corresponding risk attribute (such as "Metal Impurities") from the inherited attribute list is automatically significantly reduced, thus simulating the "purified" material state in the formulation analysis structure. Conversely, if the process parameters do not meet the effective thresholds (e.g., the user-set reaction time is too short or the temperature is insufficient), the process capability is deemed ineffective, and the risk attribute cannot be blocked. In this case, the upstream risk attribute will penetrate the process node and remain in the generated downstream material nodes. More importantly, the system will capture this abnormal logic and automatically generate a potential failure mode marker under the corresponding node in the formula analysis structure tree (e.g., automatically marking "failure mode: residual impurities" or "incomplete reaction"). This mechanism realizes automatic deduction and early warning from "improper process parameter settings" to "potential problems in formula quality" without the need for manual correlation analysis.

[0059] In the above embodiments, by introducing a comparison mechanism between process capability definitions and process parameter thresholds, the system no longer merely transmits data statically, but dynamically determines the retention or removal of risk attributes based on the effectiveness of process settings. This means the system can accurately simulate the effects of actual physicochemical processes such as "purification," "impurity removal," or "modification," and automatically block the downstream transmission of upstream risks when process parameters meet the standards. This mechanism greatly improves the realism of the formulation analysis structure, avoids misjudgments of risks due to ignoring the purification effect of processes, and ensures a high degree of consistency between the downstream material state and the actual production logic. Furthermore, when process parameters fail to meet the effective thresholds, the system can not only identify the failure of process capability, but also automatically deduce the specific impact of this failure on material properties (such as impurity penetration), and directly instantiate it as a potential failure mode marker in the formulation analysis structure. This function fundamentally solves the problem of weak correlation between PFMEA (process analysis) and MaFMEA (design analysis) in traditional FMEA, ensuring that formulation risks caused by improper parameter settings at the manufacturing end (such as incomplete reaction or impurity residue) can be automatically captured and warned by the system, effectively preventing the omission of such hidden risks during the R&D stage.

[0060] In one embodiment, the step of performing failure mode and effects analysis based on MaFMEA data of process materials includes:

[0061] In response to the parameter fluctuation identifier of the process node in the process analysis structure, the process node is identified as the cause of failure.

[0062] Based on the directed lines in the composite analysis diagram, locate the downstream material node directly pointed to by the process node corresponding to the cause of failure;

[0063] The causal mapping rule base is invoked, and the corresponding product characteristic variation results are retrieved using the parameter fluctuation identifier and the material node type as indexes.

[0064] In the formula analysis structure, the product characteristic variation results are instantiated as failure modes, and a correlation relationship is established from the failure cause to the failure mode, forming a failure chain.

[0065] Specifically, firstly, when a user marks a parameter fluctuation flag (e.g., "heating time") for a specific process node (e.g., "heating too fast") in the process analysis structure interface, this operation is triggered, locking that process node and its parameter fluctuation state as the failure cause for this analysis. Then, the data structure of the synthetic analysis graph is read, and the process is traced downstream from the locked failure cause node along predefined directed lines (i.e., reaction flow direction) to precisely locate the downstream material node (e.g., the "porous carbon" intermediate) directly pointed to by that process node. This step utilizes the connectivity of the directed acyclic graph to ensure that the failure propagation path conforms to the actual physical / chemical reaction sequence. Next, a pre-built causal mapping rule base is invoked, using the captured "parameter fluctuation flag" (heating too fast) and the located "material node type" (porous carbon) as a joint index key for retrieval in the database. If a matching item exists in the rule base, the system will extract the product characteristic variation results recorded in the database (e.g., "uneven pore size distribution" or "structural collapse"). Finally, it will jump to the formulation analysis structure, find the corresponding material node, and instantiate the retrieved product characteristic variation result into a specific failure mode for that node. Simultaneously, a logical relationship is automatically established in the database, pointing from the "failure cause in the process analysis structure" to the "failure mode in the formulation analysis structure," thus forming a complete and visualized failure chain. This process eliminates the need for manual, complex cross-level connections, achieving automatic causal deduction from process fluctuations to potential formulation quality issues.

[0066] Taking the preparation process of hard carbon anode materials as an example, the automatic pathfinding and construction process of the failure chain described above is explained in detail. Assume that the "synthesis analysis diagram" defines a path from the "precursor" through "OP10 carbonization and calcination" to generate a "porous carbon intermediate," ultimately yielding the "hard carbon finished product." When researchers mark the "heating rate" parameter as "too fast" under the "OP10 carbonization and calcination" node in the process analysis structure, the system immediately identifies "too fast heating rate" as the cause of failure. Subsequently, the system traces downstream based on the directed lines in the synthesis analysis diagram, locating the downstream material node directly affected by the calcination process as the "porous carbon intermediate." The system then calls the pre-set causal mapping rule base in the background, using "carbonization and calcination," "heating rate," "too fast," and "porous carbon" as joint indexes for retrieval, matching the corresponding rule entry: "Too fast heating leads to pore structure collapse or uneven distribution." Based on this search result, the system automatically instantiates a failure mode with the content "uneven pore distribution" under the "porous carbon intermediate" node in the formulation analysis structure. Next, the system continues to search down the topological path to the final product "hard carbon" node, and according to the logic in the rule base that "uneven pore distribution leads to a decrease in specific capacity," it generates the failure consequence "battery specific capacity does not meet standards" at the product end. Finally, the system automatically generates and displays a visual failure chain on the interface: [Process failure: calcination heating rate too fast] → [Formulation failure: uneven pore distribution in intermediates] → [Product failure: hard carbon specific capacity does not meet standards]. R&D personnel can directly see the specific impact path of process fluctuations on the final product performance without manually searching and connecting across levels.

[0067] In this embodiment, taking hard carbon anode material as an example, the system can automatically correlate the macroscopic process parameter of "heating rate" with the microscopic material characteristic of "pore structure," thereby deriving the final product performance index of "specific capacity." This cross-level automatic deduction mechanism breaks down the human barriers between process and formulation analysis in traditional FMEA, ensuring a logical closed loop from fluctuations at the manufacturing end to potential quality issues at the R&D end. It also significantly reduces the complexity of the analysis work and improves the reliability of the results. Faced with hundreds or thousands of process parameters in the production of process materials, manually identifying their impact on downstream product characteristics is not only extremely labor-intensive but also prone to errors. This solution, by solidifying an expert rule base (e.g., the rule "excessive heating leads to pore collapse"), achieves the digital inheritance and reuse of expert experience. This not only significantly reduces repetitive work for engineers but, more importantly, effectively prevents the omission of serious failure modes due to insufficient analyst experience, significantly improving the depth, accuracy, and reference value of FMEA analysis reports.

[0068] In one embodiment, the causal mapping rule base contains multiple sets of key-value pair data, wherein:

[0069] Key elements include process type, process parameter type, and fluctuation direction;

[0070] The values ​​include the physical and / or chemical properties of the affected materials and their variation trends;

[0071] The steps for establishing the failure chain include:

[0072] Determine whether the failure mode in the failure chain is a failure of the final product;

[0073] If not, then the failure mode described above is taken as the new cause of failure and deduced to the final product node.

[0074] Specifically, the causal mapping rule base employs a highly structured key-value pair data model. The "key" is defined by three dimensions: process type (e.g., "carbonization and roasting"), process parameter type (e.g., "heating rate"), and fluctuation direction (e.g., "too fast"), used to precisely pinpoint the triggering conditions for failure. The "value" defines the physical or chemical properties (e.g., "pore structure") and their variation trends (e.g., "collapse" or "closure") of the affected material directly caused by the triggering condition. This structured data storage method ensures that chemical reaction mechanisms can be quickly retrieved and matched by computer algorithms. Based on this, a recursive deduction algorithm can be used when establishing failure chains. After completing the first-level deduction, mapping process parameter fluctuations to the failure modes of downstream materials (e.g., intermediates), the endpoint determination logic is automatically executed: determining whether the current material node is the root node (i.e., the final product) in the synthesis analysis diagram. If the determination result is "no" (i.e., only the intermediate has failed), the failure mode of the intermediate (e.g., "intermediate pore structure collapse") will be transformed into a new failure cause for the next level of reaction. Subsequently, the steps of locating downstream nodes and searching the rule base are repeated to query the impact of "pore structure collapse" on the characteristics of the next-level product. This cyclical process will continue along the path of the directed acyclic graph, progressing layer by layer until it reaches the final product node (such as "hard carbon finished product"), thereby constructing a complete failure causal chain that runs through "process parameters - intermediate microstructure - final product macroscopic performance".

[0075] In this embodiment, firstly, by defining key-value pairs in a structured manner, complex chemical reaction experience is transformed into computable logical rules, ensuring the accuracy of the deduction. Secondly, in complex chemical processes, quality defects in intermediates (such as microscopic porosity issues) are often difficult to detect directly, and their consequences usually only become apparent in the final product (such as battery capacity). This solution, through automated multi-level recursive deduction, links the microscopic hidden dangers in the intermediate process to the macroscopic failures of the final product that are perceptible to the customer, ensuring the depth and penetration of risk analysis and effectively preventing the break and omission of key quality risks.

[0076] Furthermore, in one embodiment, the key-value pair data in the causal mapping rule base is generated based on a dual-channel knowledge extraction mechanism, including:

[0077] Collect historical production batch data of process materials, align the time-series parameters of the process with the quality inspection data of the final product, use multidimensional correlation analysis algorithms to mine the statistical impact weight of process parameter fluctuations on material properties, and generate a data-driven rule candidate set.

[0078] We construct a vertical large language model for the chemical industry, perform unstructured text analysis on historical FMEA reports, process mechanism documents and industry technical literature, extract causal entity relationships of "parameter anomaly - intermediate defect - product failure", and generate a knowledge-driven rule candidate set.

[0079] Specifically, the key-value pair data in the causal mapping rule base is not solely dependent on manual input, but is automatically generated through a knowledge extraction mechanism. This mechanism aims to integrate objective real-world data with subjective expert knowledge to construct a high-quality rule base.

[0080] First, historical production data for process materials is collected in batches, including time-series process parameters and corresponding final product inspection data from the quality management system. Addressing the unique "time lag" characteristic of process industries, batch alignment is performed, mapping the final quality of a specific batch of products back to the process parameters at its production moment, forming a one-to-one structured training dataset. Subsequently, a multidimensional correlation analysis algorithm is used to extract the statistical impact weights of process parameter fluctuations on material properties from the massive dataset. For example, the algorithm might discover a strong negative correlation between the "calcination stage heating rate" and the "final product specific surface area," thereby generating a statistically significant data-driven rule candidate entry.

[0081] Simultaneously, a large language model fine-tuned with chemical industry corpus is deployed. This model possesses the ability to understand complex chemical reaction mechanism descriptions and can perform deep semantic analysis on unstructured text data accumulated within the company, including historical FMEA analysis reports, process mechanism specifications, and even external industry technical documents. Through natural language processing technology, the large model accurately extracts causal entity relationships of "parameter anomaly - intermediate defect - product failure" from the text. For example, it identifies the logical description "If the pickling time is insufficient (parameter anomaly), it will lead to incomplete removal of metal ions (intermediate defect), which in turn causes a high battery self-discharge rate (product failure)" from a process document and transforms it into structured knowledge-driven rule candidate entries.

[0082] In the above embodiments, the dual verification mode greatly improves the accuracy and robustness of the rule base. Data mining ensures that the rules conform to the actual production situation, while large model extraction ensures that the rules conform to the principles of chemical reactions. The two complement each other, providing a solid and reliable knowledge foundation for automated failure chain deduction.

[0083] Furthermore, in one embodiment, the generation of the key-value pair data further includes a fusion instantiation step of the data-driven rule candidate set and the knowledge-driven rule candidate set:

[0084] By logically comparing the rules in the two candidate sets, strong association rules that have both statistical significance and logical rationality are selected.

[0085] When two candidate sets conflict, knowledge-driven rules or marking them as pending manual confirmation are preferred based on preset confidence weights.

[0086] The validated rules are standardized into key-value pair structures containing "process type-parameter-fluctuation" as the key and "material-characteristic-variation" as the value, and stored in the causal mapping rule base.

[0087] Specifically, firstly, the cross-validation engine is activated to logically compare the data-driven and knowledge-driven rule candidate sets. The "intersection" between the two sets is then identified, selecting rules that exhibit strong statistical correlation in historical production data and have clear causal descriptions in industry mechanistic documentation (e.g., extracted with high confidence by a large model). For inconsistencies between the two candidate sets, a conflict resolution strategy is implemented. When data mining shows a high correlation between two parameters, but the large model finds no supporting evidence in the mechanistic documentation (possibly due to data noise or spurious correlations), or when the large model extracts a theoretical rule but it doesn't show a significant correlation in actual production data (possibly due to excellent process control preventing fluctuations), a decision is made based on preset confidence weights. Typically, knowledge-driven rules are prioritized to ensure logical consistency with chemical principles; alternatively, conflicting rules are marked as "awaiting manual confirmation" and pushed to expert users for review to prevent erroneous rules from being included. Finally, all validated or confirmed rules are formatted into a unified key-value pair structure. Process type, process parameters, and fluctuation direction are combined and encapsulated into unique keys, while affected materials, characteristics, and variation trends are encapsulated into corresponding values. These standardized key-value pairs are ultimately stored in the system's causal mapping rule base, serving as the underlying logical foundation for subsequent automated FMEA analysis.

[0088] For example, by extracting the "process type" (e.g., "OP10 carbonization roasting"), specific "process parameters" (e.g., "heating rate"), and the "fluctuation direction" of these parameters (e.g., "too fast"), a unique index key is formed that can accurately pinpoint a specific abnormal state in the process. Correspondingly, the physical or chemical consequences of this triggering condition are encapsulated as a corresponding value. In the hard carbon preparation scenario described above, this value includes the "affected materials" (e.g., "porous carbon intermediates") directly affected by this process fluctuation, the changed "properties" (e.g., "pore structure"), and the specific "variation trend" (e.g., "pore size collapse or uneven distribution"). This encapsulation method transforms complex chemical reaction mechanisms into computer-readable structured data, clearly defining the causal logic from process fluctuations to changes in material properties.

[0089] Furthermore, in one embodiment, the causal mapping rule base is also configured with a dynamic weight calibration mechanism, including:

[0090] Continuously collect analysis data on non-conforming products or feedback data on after-sales failures in actual production;

[0091] The actual failure cases are compared with the existing key-value pairs in the rule base. If the actual case verifies a rule, the confidence weight of the variation trend in that rule is increased.

[0092] If the actual case does not conform to the existing rules, the process of correcting or adding rules to the rule base is triggered to achieve adaptive iteration of causal mapping logic.

[0093] Specifically, the system continuously collects analysis data on non-conforming products and failure feedback data from actual situations. Based on this, a comparison and verification algorithm is executed periodically. The collected actual failure cases are matched and analyzed against existing key-value pairs in the rule base. If an actual case verifies the accuracy of a rule (e.g., actual data confirms that "high calcination temperature" indeed leads to "decrease in specific capacity"), the confidence weight of that rule in the library will be automatically increased. This means that in future automated pathfinding simulations, this rule will have a higher priority for recommendation. Conversely, if an actual case does not conform to existing rules (e.g., a new unknown failure occurs, or the actual consequences are drastically different from the rule predictions), a process for revising or adding rules to the rule base will be triggered. For unknown failure modes, they are marked as "newly discovered rules" and pushed to experts for review and inclusion in the library; for conflicting rules, the weight of the original rule is reduced or manual correction is prompted. Through this mechanism, the causal mapping logic achieves adaptive iteration as production experience accumulates.

[0094] In the above embodiments, firstly, during the rule generation stage, a rigorous logical filter is established through dual cross-validation using both "data-driven" and "knowledge-driven" approaches, effectively addressing the limitations of a single technical path. On one hand, the theoretical support from mechanistic documents effectively eliminates "spurious correlation" noise commonly found in simple data mining—that is, situations where data is statistically correlated but lacks actual chemical causal relationships. On the other hand, empirical evidence from actual production data compensates for the potential for purely theoretical derivations to deviate from the actual technological status quo. Furthermore, the confidence level adjudication and manual review strategies adopted for conflicting rules further ensure that the rules included in the database must conform to the basic principles of chemical reactions, providing a reliable knowledge base for subsequent automated FMEA analysis.

[0095] Secondly, this embodiment standardizes and structures knowledge by uniformly formatting verified rules into a key-value pair structure containing "process-parameter-fluctuation" as the key and "material-characteristic-variation" as the value. This processing method transforms complex unstructured chemical engineering knowledge into structured data that computer algorithms can directly access, greatly optimizing the retrieval performance during subsequent automatic failure chain pathfinding. This allows the system to maintain millisecond-level response speeds even when dealing with massive amounts of process parameters, meeting the real-time requirements of industrial software and significantly improving the computational efficiency of automated analysis.

[0096] Finally, by introducing a dynamic weight calibration mechanism, we can use real-world failure cases to reverse-verify and correct the rule base, thereby improving the recommendation weight of failed rules.

[0097] In one embodiment, the method further includes, when constructing MaFMEA data for process materials:

[0098] Extract the reaction parameters defined in the process nodes of the synthesis analysis diagram and mark them as safety baseline values;

[0099] Monitor the changes in the set values ​​of the process nodes in the process analysis structure;

[0100] When a change in the set value is detected, the deviation between the changed set value and the safety benchmark value is calculated.

[0101] Determine whether the deviation value is within the preset allowable tolerance range;

[0102] If the tolerance is exceeded, a warning will be generated.

[0103] Specifically, in constructing the MaFMEA data for process materials, the reaction parameters defined at each process node in the synthesis analysis diagram (such as upper limit of reaction temperature, minimum reaction time, etc.) are first extracted and marked as safety baseline values. Then, a real-time monitoring program is initiated in the process analysis structure to continuously monitor the set values ​​of each corresponding "process node" (i.e., the proposed parameters in the actual production process card). When a user modifies the set value of a process node in the process analysis structure, a verification logic is immediately triggered. The deviation between the changed set value and the aforementioned safety baseline value is automatically calculated and compared with the preset tolerance range. If the calculated deviation exceeds the tolerance range, the current process setting is determined to violate the upstream defined reaction mechanism, and a significant warning is generated on the visualization interface. This warning can manifest as the input box turning red or automatically inserting a high-risk failure record into the FMEA table, forcing engineers to reassess the rationality of the parameter setting.

[0104] In the above embodiments, by establishing the reaction parameters in the synthesis analysis chart as safety baseline values ​​and monitoring process settings for deviations in real time, this solution effectively prevents production departments from arbitrarily changing key process parameters (such as shortening reaction time or lowering temperature without authorization) in pursuit of efficiency or cost reduction. This avoids hidden quality accidents caused by such changes and ensures that the manufacturing process always operates within the safe window allowed by the chemical mechanism. Furthermore, this real-time visual warning significantly improves the immediacy and effectiveness of risk management. This mechanism not only reduces the cost of subsequent rectification but also compels collaborative communication between R&D and process personnel, ensuring that every process change deviating from the baseline has undergone thorough risk assessment and confirmation.

[0105] Furthermore, some embodiments of this application also provide an electronic device. The electronic device can be various forms of digital computer, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, etc. The electronic device can also be various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices.

[0106] The electronic device includes: one or more processors; and a memory storing computer program instructions that, when executed, cause the processor to perform the steps of the methods provided in any one or more of the above embodiments. Figure 5An exemplary structural diagram of the electronic device is disclosed. The electronic device includes one or more processors 1101, a memory 1102, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components are interconnected via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the electronic device, including instructions stored in or on memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface). In some other embodiments, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple electronic devices can be connected, each providing some of the necessary operations. The components, their connections and relationships, and their functions shown herein are merely examples and are not intended to limit the implementation of the present application described and / or claimed herein.

[0107] The electronic device may further include an input device 1103 and an output device 1104. The processor 1101, memory 1102, input device 1103 and output device 1104 may be connected by a bus or other means, as shown in the figure, which is connected by a bus.

[0108] Input device 1103 can receive input numerical or character information, and generate key signal inputs related to user settings and function control of the electronic device, such as a touch screen, keypad, mouse, trackpad, touchpad, joystick, one or more mouse buttons, trackball, joystick, etc. Output device 1104 may include a display device, auxiliary lighting device (e.g., LED), and haptic feedback device (e.g., vibration motor). The display device may include, but is not limited to, a liquid crystal display, a light-emitting diode display, and a plasma display. In some embodiments, the display device may be a touch screen.

[0109] To provide interaction with the user, the electronic device can be a computer. The computer has: a display device (e.g., a cathode ray tube or LCD monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback); and input from the user can be received in any form (e.g., voice input or tactile input).

[0110] In this embodiment, a computer-readable medium stores a computer program / instructions that, when executed by a processor, implement the steps of the methods provided in any one or more of the above embodiments. This computer-readable medium may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into that device. The aforementioned computer-readable medium carries one or more computer-readable instructions.

[0111] The memory 1102 can serve as a non-transitory computer-readable storage medium, used to store non-transitory software programs, non-transitory computer-executable programs, and modules. The processor 1101 executes various functional applications and data processing of the server by running the non-transitory software programs, instructions, and modules stored in the memory 1102, thereby implementing the program instructions / modules corresponding to the methods provided in any one or more of the embodiments described above in this application.

[0112] The memory 1102 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the electronic device. Furthermore, the memory 1102 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 1102 may optionally include memory remotely located relative to the processor 1101, and these remote memories can be connected to the electronic device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0113] It should be noted that the computer-readable medium described in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. Computer-readable media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, electrical connections having one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0114] Computer-readable media include permanent and non-permanent, removable and non-removable media, which can store information by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory, static random access memory, dynamic random access memory, other types of random access memory, read-only memory, electrically erasable programmable read-only memory, flash memory or other memory technologies, read-only optical discs, digital versatile optical discs or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0115] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including local area networks (LANs) or wide area networks (WANs), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0116] In the above embodiments, all or part of the implementation can be achieved through software, hardware, firmware, or any combination thereof. For example, it can be implemented using an application-specific integrated circuit (ASIC), a general-purpose computer, or any other similar hardware device. In some embodiments, the software program of this application can be executed by a processor to implement the above steps or functions. Similarly, the software program of this application (including related data structures) can be stored in a computer-readable recording medium, such as RAM memory, magnetic or optical drives, floppy disks, and similar devices. In addition, some steps or functions of this application can be implemented in hardware, for example, as circuitry that cooperates with a processor to perform the various steps or functions.

[0117] The computer program product provided in this application includes one or more computer programs / instructions. When executed by a processor, these computer programs / instructions generate, in whole or in part, the processes or functions described in this application. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0118] The flowcharts or block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of devices, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-specific system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0119] The scope of this application is defined by the appended claims rather than the foregoing description, and is therefore intended to encompass all variations falling within the meaning and scope of equivalents of the claims. No reference numerals in the claims should be construed as limiting the scope of the claims. Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a device claim may also be implemented by a single unit or device in software or hardware. Terms such as "first," "second," etc., are used only for distinguishing descriptions and do not indicate any particular order, nor should they be construed as indicating or implying relative importance.

[0120] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily made by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims, and the above embodiments should be regarded as exemplary and non-limiting.

Claims

1. A MaFMEA analysis method for process materials, characterized in that, Includes the following steps: Construct a synthetic analysis graph, which includes material nodes and process nodes, and the material nodes and process nodes are connected by directed lines to form a directed acyclic graph structure; Traverse the synthesis analysis graph, extract the material nodes in the synthesis analysis graph, establish the parent-child inheritance logic between materials based on the pointing relationship of the directed connection in the directed acyclic graph, and construct a hierarchical formula analysis structure. Traverse the synthetic analysis graph, extract process nodes from the synthetic analysis graph, and construct a time-series process analysis structure based on the topological timing logic of the process nodes in the directed acyclic graph. The reaction parameters and characteristic indicators defined at each node in the synthesis analysis diagram are mapped to the corresponding nodes in the process analysis structure and the formulation analysis structure, respectively, to construct process material MaFMEA data; Failure Mode and Effects Analysis was performed based on the MaFMEA data of the process materials.

2. The method according to claim 1, characterized in that, The step of mapping the reaction parameters and characteristic indices defined at each node in the synthesis analysis diagram to the corresponding nodes in the process analysis structure and the formulation analysis structure further includes: Configure material type labels for the material nodes, and map the material nodes to the corresponding classification levels in the recipe analysis structure based on the material type labels; Configure element attribution attributes for the reaction constraints of the process nodes, and map the reaction constraints to the corresponding production element categories in the process analysis structure based on the element attribution attributes.

3. The method according to claim 1, characterized in that, The steps of traversing the synthesis analysis graph, extracting material nodes from the synthesis analysis graph, establishing parent-child inheritance logic between materials based on the pointing relationships of directed connections in the directed acyclic graph, and constructing a hierarchical formula analysis structure further include: Define risk attributes for the material node, including impurity content, by-product risk, or physical property parameters; When establishing the parent-child inheritance logic between materials, the risk attributes of upstream material nodes are inherited to downstream material nodes according to the topological relationship of directed connections, forming an attribute transfer flow.

4. The method according to claim 3, characterized in that, When establishing the parent-child inheritance logic between materials, the step of passing on the risk attributes of upstream material nodes to downstream material nodes to form an attribute transfer flow, based on the topological relationship of directed connections, further includes: When the attribute transfer flow passes through the process node, the process capability definition and corresponding process parameters of the process node are read. If the process parameters meet the effective threshold in the process capability definition, the corresponding risk attribute is reduced when generating downstream material nodes; If the process parameters do not meet the effective threshold, the process capability is determined to be in failure, and the corresponding risk attribute is maintained when generating downstream material nodes, and a potential failure mode marker is generated in the formula analysis structure.

5. The method according to claim 1, characterized in that, The steps for performing failure mode and effects analysis based on MaFMEA data for process materials include: In response to the parameter fluctuation identifier of the process node in the process analysis structure, the process node is identified as the cause of failure. Based on the directed lines in the composite analysis diagram, locate the downstream material node directly pointed to by the process node corresponding to the cause of failure; The causal mapping rule base is invoked, and the corresponding product characteristic variation results are retrieved using the parameter fluctuation identifier and the material node type as indexes. In the formula analysis structure, the product characteristic variation results are instantiated as failure modes, and a correlation relationship is established from the failure cause to the failure mode, forming a failure chain.

6. The method according to claim 5, characterized in that, The causal mapping rule base contains multiple sets of key-value pair data, wherein: Key elements include process type, process parameter type, and fluctuation direction; The values ​​include the physical and / or chemical properties of the affected materials and their variation trends; The steps for establishing the failure chain include: Determine whether the failure mode in the failure chain is a failure of the final product; If not, then the failure mode described above is taken as the new cause of failure and deduced to the final product node.

7. The method according to claim 1, characterized in that, The method further includes, when constructing MaFMEA data for process materials: Extract the reaction parameters defined in the process nodes of the synthesis analysis diagram and mark them as safety baseline values; Monitor the changes in the set values ​​of the process nodes in the process analysis structure; When a change in the set value is detected, the deviation between the changed set value and the safety benchmark value is calculated. Determine whether the deviation value is within the preset allowable tolerance range; If the tolerance is exceeded, a warning will be generated.

8. An electronic device, characterized in that, The electronic device includes: One or more processors; and A memory storing computer program instructions, which, when executed, cause the processor to perform the steps of the method as described in any one of claims 1 to 7.

9. A computer-readable medium having a computer program / instructions stored thereon, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 7.

10. A computer program product comprising a computer program / instructions, characterized in that, When the computer program / instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 7.

Images