Cigarette production quality risk management and control method, device and equipment

Abstract

The invention discloses a cigarette production quality risk management and control method, device and equipment, and relates to the technical field of tobacco quality management, and the method comprises the steps: obtaining cigarette production process flow data; performing failure mode identification on the cigarette production process flow data to obtain potential failure mode data; determining a risk coping object according to the potential failure mode data; and determining production process improvement measures based on the cigarette production knowledge graph according to the risk response object. According to the scheme, advanced early warning and accurate intervention of quality risks can be achieved, and the standardized control level and the product quality stability of the cigarette production process are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of tobacco quality management, and in particular to a method, device and equipment for controlling quality risks in cigarette production. Background Art

[0002] Quality risk management is to reduce the existence of risk sources and the probability of risk events by identifying, evaluating and controlling a series of factors that cause product quality risks, thereby reducing risk losses and keeping risk losses at an acceptable level for decision makers.

[0003] Current quality risk management only investigates and controls single potential quality risks and is unable to match and integrate multi-dimensional data, resulting in low production quality control levels and unstable product quality. Summary of the Invention

[0004] In view of the above, the present invention aims to provide a method, device and equipment for controlling cigarette production quality risks to solve the above-mentioned technical problems.

[0005] The technical solution adopted in the present invention is as follows:

[0006] The present invention provides a method for controlling cigarette production quality risks, which includes:

[0007] Obtain cigarette production process data;

[0008] Performing failure mode identification on the cigarette production process flow data to obtain potential failure mode data;

[0009] Determine risk response targets based on the potential failure mode data;

[0010] According to the risk response objects, production process improvement measures are determined based on the cigarette production knowledge graph.

[0011] Optionally, performing failure mode identification on the cigarette production process flow data to obtain potential failure mode data includes:

[0012] Dividing the cigarette production process into a plurality of sub-processes according to the cigarette production process flow data;

[0013] Failure mode identification is performed on each sub-process according to preset conditions to obtain potential failure mode data of each sub-process, wherein the potential failure mode data at least includes failure severity, failure frequency and failure detectability.

[0014] Optionally, determining risk response targets based on the potential failure mode data includes:

[0015] determining a risk index based on the failure mode data;

[0016] According to the risk index, the risk response object is determined.

[0017] Optionally, determining a risk index based on the failure mode data includes:

[0018] The risk index is obtained by multiplying the failure severity, failure frequency and failure detectability.

[0019] Optionally, determining a risk response target based on the risk index includes:

[0020] Construct failure cause correlation matrix;

[0021] Determining a risk object level according to the failure cause association matrix and the risk index;

[0022] Sorting the risk object levels according to preset rules to obtain a priority sequence of risk objects;

[0023] Identify the highest priority risk objects as risk response targets.

[0024] Optionally, according to the risk response object, production process improvement measures are determined based on the cigarette production knowledge graph, including:

[0025] Constructing a knowledge graph for cigarette production, wherein the knowledge graph includes production process improvement measures;

[0026] Mapping the risk index to the cigarette production knowledge graph;

[0027] According to the mapping relationship between the risk index and the cigarette production knowledge graph, the production process improvement measures corresponding to the risk response object are determined.

[0028] Optionally, build a knowledge graph for cigarette production, including:

[0029] Obtaining cigarette production data, the cigarette production data including at least production equipment logs, process material parameters, and historical quality inspection data;

[0030] The cigarette production data is parsed and processed using a BiLSTM-CRF model to obtain a cigarette production knowledge graph.

[0031] The present invention also provides a device for controlling cigarette production quality risks, comprising:

[0032] An acquisition module is used to obtain cigarette production process data;

[0033] The processing module is used to perform failure mode identification on the cigarette production process flow data to obtain potential failure mode data; determine risk response objects based on the potential failure mode data; and determine production process improvement measures based on the cigarette production knowledge graph according to the risk response objects.

[0034] The present invention further provides a computing device, comprising: a processor and a memory storing a computer program, wherein the computer program executes the method described above when executed by the processor.

[0035] The present invention also provides a computer-readable storage medium storing instructions, which, when executed on a computer, enable the computer to execute the method described above.

[0036] The above solution of the present invention includes at least the following beneficial effects:

[0037] The above-mentioned solution of the present invention obtains cigarette production process data; performs failure mode identification on this data to obtain potential failure mode data; determines risk response targets based on this potential failure mode data; and, based on these risk response targets, determines production process improvement measures based on the cigarette production knowledge graph. This solution of the present invention can achieve advanced warning and precise intervention of quality risks, significantly improving the standardization control level of the cigarette production process and product quality stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described below with reference to the accompanying drawings, in which:

[0039] Figure 1 This is a flow chart of a method for controlling cigarette production quality risks provided by an embodiment of the present invention.

[0040] Figure 2 This is a fusion framework diagram of the standard operating procedure system based on potential failure modes provided by an embodiment of the present invention.

[0041] Figure 3 A schematic diagram of a module of a device for controlling cigarette production quality risks provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0042] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.

[0043] The present invention proposes an embodiment of a method for controlling cigarette production quality risks. Specifically, Figure 1 shown, including:

[0044] Step 11, obtaining cigarette production process data;

[0045] Step 12, performing failure mode identification on the cigarette production process flow data to obtain potential failure mode data;

[0046] Step 13: determining risk response targets based on the potential failure mode data;

[0047] Step 14: Determine production process improvement measures based on the cigarette production knowledge graph according to the risk response object.

[0048] In this example, the scope of the cigarette production process should include all production steps, from raw materials leaving the cabinet to cigarette packaging. This can be demonstrated by drawing a process flow chart for the cigarette production process. A process flow chart includes process outputs (product characteristics, requirements, deliverables, etc.) and inputs (process characteristics, sources of variation, etc.), as well as each step of the cigarette production process. This flow chart can identify each process related to cigarette production and its corresponding function, the impact of each step on product quality, and enable a preliminary assessment of quality risks.

[0049] The cigarette production process flow data is used to identify the functions of each link and process in the entire cigarette manufacturing process, such as silk making, packaging, filter rod forming, power workshops, as well as raw material and auxiliary material supply, finished product warehousing, etc. According to the process operation instructions, a brief description of the analyzed process or process is given, and the purpose of the process or process is explained as simply as possible. On the basis of the functional definition, the analysis focuses on identifying the potential failure modes of each function under specified conditions, and a comprehensive list and definition of failure modes in the cigarette manufacturing process are generated according to the cigarette product quality formation path.

[0050] Based on the cigarette manufacturing process failure mode list generated in step 12, potential failure mode consequences are determined, and the failure cause of each failure mode is determined using the failure cause association matrix. Furthermore, the failure severity, failure frequency, and failure detection of the defects are analyzed to identify the object with the highest risk index and determine it as the risk response target. Finally, using the cigarette production knowledge graph, response and improvement measures are automatically matched to this risk response target.

[0051] The cigarette production quality risk management method of this embodiment can automatically identify and predict potential quality risks in the entire process of silk making, packaging, power, logistics, etc., build a process knowledge base through knowledge graph technology, associate multi-dimensional data such as equipment, parameters, and materials, and realize intelligent matching and recommendation of failure modes; at the same time, it dynamically predicts the risk evolution trend and adaptively optimizes control measures, thereby establishing a systematic and intelligent quality prevention and control system.

[0052] In an optional embodiment of the present invention, step 12 may include:

[0053] Step 121, dividing the cigarette production process into a plurality of sub-processes according to the cigarette production process flow data;

[0054] Step 122 : performing failure mode identification on each sub-process according to preset conditions to obtain potential failure mode data of each sub-process. The potential failure mode data at least includes failure severity, failure frequency, and failure detectability.

[0055] In this example, the cigarette production process is divided into multiple sub-processes, including various steps and processes in the entire cigarette manufacturing process, such as thread making, packing, filter rod forming, power generation, raw material and auxiliary material supply, and finished product warehousing. The functions of each step and process in the entire cigarette manufacturing process are identified. Based on the process operation instructions, a brief description of the analyzed process or process is provided, along with the purpose of the process or process explained as simply as possible. If the production process includes multiple steps with different failure modes, these steps can be listed as independent processes.

[0056] Based on the functional definition, the focus is on analyzing the phenomena such as the inability to complete the established functions under specified conditions (environment, operation, time), the inability to maintain product parameter values ​​between the specified upper and lower limits, the product not meeting the national standards, or the product being within the specified range, resulting in reduced customer satisfaction with cigarette products. The potential failure modes of each function are identified, and a comprehensive list and definition of failure modes in the cigarette manufacturing process are generated according to the cigarette product quality formation path.

[0057] In an optional embodiment of the present invention, step 13 may include:

[0058] Step 131, determining a risk index based on the failure mode data;

[0059] Step 132: Determine the risk response target based on the risk index.

[0060] In this embodiment, risk response targets are determined by the consequences of potential failure modes. Potential failure mode consequences are defined as the consequences of a failure mode perceived by customers. Failure consequences should be described in terms of situations that customers might notice or experience. Remember that customers can be internal customers or end consumers. Each type of customer must be considered when evaluating the consequences of potential failure modes. After analyzing the consequences of various potential failure modes, these consequences are assigned a severity rating. The severity rating is a score corresponding to the most severe consequence of a given potential failure mode.

[0061] Specifically, the severity (S), detectability (D), and frequency (O) of the potential failure modes in the list are first analyzed. For each potential failure mode, using methods such as quality function deployment, within the widest possible scope, systematically analyze the specific causes of each failure mode in each process and each procedure from the perspectives of people (personnel), machines (equipment, etc.), materials (materials, etc.), methods (working methods), environment (environment), and measurements (measurements). Simultaneously, using the failure cause correlation matrix, a cause expansion is performed on the identified failure mode causes. Based on the theoretical correlation between the failure cause and the failure mode and the actual frequency of the failure mode caused by the failure cause, the correlation between the failure cause and the failure mode is quantified to determine the main cause of the failure mode.

[0062] Frequency is the probability of failure occurring due to each causal mechanism corresponding to a potential failure mode. The probability of occurrence of potential causes of failure is evaluated on a scale of 1 to 10. For different causal mechanisms of various failure modes, the current control measures and detection levels are analyzed in detail. Current process control is a description of the methods for preventing the occurrence of potential failure modes and detecting failure modes within the possible scope of the production process. The current control measures are analyzed from two aspects: one is prevention, which is to eliminate (prevent) the occurrence of the causal mechanism of potential failure modes or the occurrence of failure modes, or to reduce the frequency of occurrence; the other is detection, which is to identify (detect) the occurrence of the causal mechanism of potential failure modes or the occurrence of failure modes, so as to lead to the development of relevant improvement measures or preventive measures.

[0063] The detectability describes the difficulty of discovering (detecting) the causal mechanism under current conditions. The easier it is to detect, the lower the detectability, and the more difficult it is to detect, the higher the detectability.

[0064] The risk level is evaluated by multiplying the severity (S), detection (O) and frequency (D) to obtain the risk index. The higher the risk index, the higher the priority of failure mode handling.

[0065] In an optional embodiment of the present invention, step 132 may include:

[0066] Step 1321, constructing a failure cause correlation matrix;

[0067] Step 1322: Determine the risk object level based on the failure cause association matrix and the risk index;

[0068] Step 1323: sort the risk object levels according to a preset rule to obtain a priority sequence of the risk objects;

[0069] Step 1324: determine the risk object with the highest priority as the risk response object.

[0070] In this embodiment, a failure cause association matrix is ​​established. First, a cause list is constructed, which includes failure causes from the dimensions of people (personnel), machine (equipment, etc.), material (materials, etc.), method (working method), and environment (environment):

[0071] Table 1. List of reasons (example):

[0072]

[0073] Next, we formulate the rules for quantifying the association strength, use a 9-point scoring system, and define the mapping function:

[0074] ;

[0075] in For the degree of correlation.

[0076] Generate a failure cause correlation matrix based on the cause list and correlation strength quantification rules (example):

[0077] Table 2. Failure cause correlation matrix:

[0078]

[0079] Rank the failure mode causes in the failure cause correlation matrix from high to low risk level. Find the failure mode cause with the highest risk score and identify it as the highest-priority risk target, thus becoming the target for risk response. Based on the actual circumstances of the tobacco industry and enterprises, research and develop risk level determination and risk optimization principles to identify risk response targets. For risk response, propose response and improvement measures, and develop a detailed implementation plan for each measure, clearly defining the responsible departments, individuals, and completion deadlines.

[0080] In an optional embodiment of the present invention, step 14 may include:

[0081] Step 141 constructs a knowledge graph of cigarette production, wherein the knowledge graph includes production process improvement measures;

[0082] Step 142: Mapping the risk index to the cigarette production knowledge graph;

[0083] Step 143: Determine the production process improvement measures corresponding to the risk response object based on the mapping relationship between the risk index and the cigarette production knowledge graph.

[0084] In this embodiment, first, cigarette production data is obtained, which includes at least production equipment logs, process material parameters, and historical quality inspection data;

[0085] The cigarette production data is parsed and processed using the BiLSTM-CRF model to obtain a knowledge graph of cigarette production.

[0086] Specifically, by integrating heterogeneous data from multiple sources, such as MES system equipment logs, process bills of materials (BOMs), and quality inspection databases, a standardized knowledge base covering the entire cigarette production process was established. This enabled structured associations between core entities such as equipment, process parameters, raw materials, and quality indicators. For example, the "cigarette maker" equipment was linked to the "cigarette weight deviation" parameter and the "cigarette paper air permeability" material property through "influence relationship" edges. Natural language processing (NLP) technology was used to automatically parse historical fault reports and SOP documents to extract failure mode relationships. A tobacco-specific dictionary was constructed based on industry terminology, and a BiLSTM-CRF model was used for entity recognition. The extracted failure modes were then integrated into the knowledge graph. Specifically, unstructured text (such as production logs) was segmented and key entities in cigarette production were annotated, such as: equipment (cigarette maker, packaging machine); materials (tobacco, filter rods); processes (tobacco cutting, rolling, packaging); and parameters (temperature (25°C) and speed (1000 cigarettes / minute). Through the BiLSTM layer in the BiLSTM-CRF model: capturing contextual semantics (for example, "rolling" is often followed by the "speed" parameter), the CRF layer: constraining the label sequence (for example, "B-PARAMETER" cannot be followed by "I-PROCESS"), then outputting entities such as equipment and processes in the production log, extracting relationships based on rules or dependency syntax analysis, and finally storing the entities and relationships in the graph database to generate a knowledge graph.

[0087] like Figure 2 As shown, in an optional embodiment of the present invention, the method for controlling cigarette production quality risks also includes: establishing a time series prediction model for key equipment parameters (such as cigarette machine blade temperature, packaging machine negative pressure value, flavoring machine flow rate, etc.) to provide early warning of the risk of parameters exceeding process control limits.

[0088] Specifically, high-frequency time series data (1s sampling) from equipment sensors is collected, and key parameters (such as the 20 core variables that affect cigarette weight) and the TCN-Attention hybrid model are screened in combination with the cigarette production knowledge graph to enhance the capture of long-term dependencies (such as the lag effect of the drying drum temperature); Bayesian optimization is used to adjust hyperparameters (number of layers, dropout rate).

[0089] A comprehensive risk index is calculated based on LSTM prediction results, failure mode correlations within the cigarette production knowledge graph, and equipment health scores (e.g., remaining life predictions). The system dynamically adjusts the priority of SOP (Standard Operating Procedure) control measures based on real-time risk levels. A decision tree engine is embedded in the MES system, allowing SOP steps to be restructured in real time based on risk levels (e.g., skipping non-critical calibration steps when risk is high and prioritizing safety shutdown checks).

[0090] While using potential failure modes to conduct failure mode and effect analysis in the cigarette product production process, the key characteristics and control methods identified in the potential failure modes are reflected in the SOP. Among them, all SOPs can be summarized into three types: operation, maintenance and business. Through the SOP, the "operation steps, operation content, key points of the operation, causes / consequences of the key points, and operation pictures" are clearly defined so that the key characteristics identified in the potential failure modes can be truly controlled, and the integrated flow charts and tables are determined to form a standardized process for potential failure modes in cigarette product production. It is then refined into an operation instruction book, and the implementation steps and methods used in the cigarette industry are standardized to form a standard application process and precautions, providing guidance for the correct application of quality management personnel and providing assistance for continuous prevention and improvement in the future.

[0091] The method for controlling quality risks in cigarette production described in the above-mentioned embodiments of the present invention integrates potential failure modes with standard operating procedures, and can automatically identify and predict potential quality risks in the entire process of silk making, packaging, power, logistics, etc. It constructs a process knowledge base through knowledge graph technology, associates multi-dimensional data such as equipment, parameters, and materials, and realizes intelligent matching and recommendation of failure modes; at the same time, it dynamically predicts the evolution trend of risks and adaptively optimizes control measures, thereby establishing a systematic and intelligent quality prevention and control system.

[0092] like Figure 3 As shown, an embodiment of the present invention further provides a cigarette production quality risk control device 30, comprising:

[0093] An acquisition module 31 is used to acquire cigarette production process data;

[0094] The processing module 32 is used to perform failure mode identification on the cigarette production process flow data to obtain potential failure mode data; determine risk response targets based on the potential failure mode data; and determine production process improvement measures based on the cigarette production knowledge graph according to the risk response targets.

[0095] Optionally, performing failure mode identification on the cigarette production process flow data to obtain potential failure mode data includes:

[0096] Dividing the cigarette production process into a plurality of sub-processes according to the cigarette production process flow data;

[0097] Failure mode identification is performed on each sub-process according to preset conditions to obtain potential failure mode data of each sub-process, wherein the potential failure mode data at least includes failure severity, failure frequency and failure detectability.

[0098] Optionally, determining risk response targets based on the potential failure mode data includes:

[0099] determining a risk index based on the failure mode data;

[0100] According to the risk index, the risk response object is determined.

[0101] Optionally, determining a risk index based on the failure mode data includes:

[0102] The risk index is obtained by multiplying the failure severity, failure frequency and failure detectability.

[0103] Optionally, determining a risk response target based on the risk index includes:

[0104] Construct failure cause correlation matrix;

[0105] Determining a risk object level according to the failure cause association matrix and the risk index;

[0106] Sorting the risk object levels according to preset rules to obtain a priority sequence of risk objects;

[0107] Identify the highest priority risk objects as risk response targets.

[0108] Optionally, according to the risk response object, production process improvement measures are determined based on the cigarette production knowledge graph, including:

[0109] Constructing a knowledge graph for cigarette production, wherein the knowledge graph includes production process improvement measures;

[0110] Associating the risk index with the cigarette production knowledge graph;

[0111] According to the mapping relationship between the risk index and the cigarette production knowledge graph, the production process improvement measures corresponding to the risk response object are determined.

[0112] Optionally, build a knowledge graph for cigarette production, including:

[0113] Obtaining cigarette production data, the cigarette production data including at least production equipment logs, process material parameters, and historical quality inspection data;

[0114] The cigarette production data is parsed and processed using a BiLSTM-CRF model to obtain a cigarette production knowledge graph.

[0115] It should be noted that the device is a device corresponding to the above method, and all implementation methods in the above method embodiments are applicable to the embodiments of the device and can achieve the same technical effects.

[0116] An embodiment of the present invention further provides a computing device comprising: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described in the above embodiment. All implementations of the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0117] The present invention also provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to execute the method described in the above embodiment. All implementations of the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0118] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present invention.

[0119] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0120] In the embodiments provided by the present invention, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interface, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0121] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0122] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0123] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard drive, ROM, RAM, a magnetic disk, or an optical disk.

[0124] In addition, it should be pointed out that in the apparatus and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present invention. Moreover, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but they do not necessarily need to be performed in chronological order, and some steps can be performed in parallel or independently of each other. For those of ordinary skill in the art, it can be understood that all or any steps or components of the method and apparatus of the present invention can be implemented in hardware, firmware, software or a combination thereof in any computing device (including a processor, storage medium, etc.) or a network of computing devices. This can be achieved by those of ordinary skill in the art using their basic programming skills after reading the description of the present invention.

[0125] Therefore, the purpose of the present invention can also be achieved by running a program or a group of programs on any computing device. The computing device can be a well-known general-purpose device. Therefore, the purpose of the present invention can also be achieved simply by providing a program product containing program code that implements the method or device. That is to say, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any well-known storage medium or any storage medium developed in the future. It should also be pointed out that in the device and method of the present invention, it is obvious that each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent schemes of the present invention. In addition, the steps of performing the above-mentioned series of processing can naturally be performed in chronological order according to the order of description, but do not necessarily need to be performed in chronological order. Certain steps can be performed in parallel or independently of each other.

[0126] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for controlling cigarette production quality risks, characterized in that: include: Obtain cigarette production process data; Performing failure mode identification on the cigarette production process flow data to obtain potential failure mode data; Determine risk response targets based on the potential failure mode data; According to the risk response objects, production process improvement measures are determined based on the cigarette production knowledge graph.

2. The method for controlling cigarette production quality risks according to claim 1, characterized in that: Failure mode identification is performed on the cigarette production process flow data to obtain potential failure mode data, including: Dividing the cigarette production process into a plurality of sub-processes according to the cigarette production process flow data; Failure mode identification is performed on each sub-process according to preset conditions to obtain potential failure mode data of each sub-process, wherein the potential failure mode data at least includes failure severity, failure frequency and failure detectability.

3. The method for controlling cigarette production quality risks according to claim 1, characterized in that: Based on the potential failure mode data, determine the risk response targets, including: determining a risk index based on the failure mode data; According to the risk index, the risk response object is determined.

4. The method for controlling cigarette production quality risks according to claim 3, characterized in that: Determine a risk index based on the failure mode data, including: The risk index is obtained by multiplying the failure severity, failure frequency and failure detectability.

5. The method for controlling cigarette production quality risks according to claim 4, characterized in that: Based on the risk index, determine the risk response targets, including: Construct failure cause correlation matrix; Determining a risk object level according to the failure cause association matrix and the risk index; Sorting the risk object levels according to preset rules to obtain a priority sequence of risk objects; Identify the highest priority risk objects as risk response targets.

6. The method for controlling cigarette production quality risks according to claim 5, characterized in that: According to the risk response targets, production process improvement measures are determined based on the cigarette production knowledge graph, including: Constructing a knowledge graph for cigarette production, wherein the knowledge graph includes production process improvement measures; Mapping the risk index to the cigarette production knowledge graph; According to the mapping relationship between the risk index and the cigarette production knowledge graph, the production process improvement measures corresponding to the risk response object are determined.

7. The method for controlling cigarette production quality risks according to claim 6, characterized in that: Construct a knowledge graph of cigarette production, including: Obtaining cigarette production data, the cigarette production data including at least production equipment logs, process material parameters, and historical quality inspection data; The cigarette production data is parsed and processed using a BiLSTM-CRF model to obtain a cigarette production knowledge graph.

8. A device for controlling cigarette production quality risks, characterized in that: include: An acquisition module is used to obtain cigarette production process data; a processing module, configured to perform failure mode identification on the cigarette production process flow data to obtain potential failure mode data; Determine risk response targets based on the potential failure mode data; According to the risk response objects, production process improvement measures are determined based on the cigarette production knowledge graph.

9. A computing device, characterized in that include: A processor and a memory storing a computer program, wherein when the computer program is executed by the processor, the method according to any one of claims 1 to 7 is performed.

10. A computer-readable storage medium, characterized in that: The device stores instructions, which, when executed on a computer, enable the computer to execute the method according to any one of claims 1 to 7.

Images