Specification operation intelligent checking and approving method based on identity dual authentication

Abstract

The present application relates to the technical field of data processing, and particularly relates to a specification operation intelligent verification and approval method based on identity dual authentication, which comprises the following steps: obtaining an operation behavior sequence by recognizing personnel action data and equipment state change data, and obtaining an operation behavior log and an operation vector based on the operation behavior sequence; determining a specification degree corresponding to an operation instruction and a stability degree corresponding to the operation instruction based on the operation behavior log; splitting a process according to the stability degree to obtain each operation segment corresponding to the process, and calculating an image weight corresponding to each operation segment; selecting a target operation segment according to the image weight, combining a time difference coefficient of an operation vector of the target operation segment corresponding to a historical reference process and the similarity of the operation vector to determine a target credibility score, and performing verification processing on the operation behavior according to the target credibility score to obtain a verification result. The present application can improve the accuracy of specification operation intelligent verification and approval.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and more specifically to a standardized intelligent verification and approval method for operations based on dual identity authentication. Background Technology

[0002] In the field of intelligent verification and approval of standardized operations, existing technologies often learn user behavior by establishing user action chains to build user action habit models in order to judge the compliance of personnel operation behaviors. For the process of users performing different actions, the differences in the overall action behavior events caused by changes in the execution results of each action are captured, and then the compliance verification of the current action sequence is completed through action sequence matching analysis.

[0003] However, existing technologies generally adopt the traditional single-point verification and comparison method, which has significant technical defects: On the one hand, this method is difficult to prevent the risk of identity replacement. When someone impersonates a legitimate person to perform an operation, the compliance performance of a single operation point cannot reflect the deviation and change of the overall process, resulting in distorted verification results. On the other hand, it cannot effectively avoid the hidden problem of single operation point compliance but abnormal operation segment. That is, a single operation step conforms to the specifications, but the behavioral characteristics of the continuous operation segment are significantly different from the person's exclusive operating habits, thereby reducing the accuracy of judging the compliance of the entire process operation and failing to meet the verification requirements of precise binding of operation behavior and actor in high-security scenarios.

[0004] In other words, the accuracy of existing technology's standardized operation verification and approval is relatively low. Summary of the Invention

[0005] To address the technical problem of low accuracy in existing technical specification operation verification and approval methods, the present invention aims to provide an intelligent verification and approval method for specification operations based on dual identity authentication. The specific technical solution adopted is as follows:

[0006] In a first aspect, one embodiment of the present invention provides a standardized intelligent verification and approval method for operations based on dual identity authentication, the method comprising:

[0007] By identifying personnel action data and equipment status change data, an operation behavior sequence is obtained, and based on the operation behavior sequence, operation behavior logs and operation vectors are acquired.

[0008] Based on operation behavior logs, the standardization and stability of operation instructions are determined. Standardization is used to characterize the degree of compliance deviation of a single operation instruction. Stability is used to characterize the long-term stable execution capability of the same operation instruction under different execution environments.

[0009] The process is broken down based on stability to obtain the corresponding operation segments, and the profile weight of each operation segment is calculated. The profile weight is used to characterize the representativeness of the operation segment to the specific operation characteristics of the personnel.

[0010] Target operation segments are selected based on profile weights. The duration difference coefficient of the operation vectors of the target operation segments corresponding to the historical benchmark process and the similarity of the operation vectors are combined to determine the target credibility score. The operation behavior is then verified based on the target credibility score to obtain the verification result.

[0011] In one embodiment, the operation behavior log includes personnel ID, operation ID, operation duration, timestamp, and process batch number. Based on the operation behavior log, the standardization of the operation instruction is determined, including:

[0012] Extract the average duration, operation duration, and percentage of execution times from the operation behavior logs;

[0013] The standardization of operation instructions is determined based on the average duration, operation duration, and percentage of executions.

[0014] In one embodiment, determining the stability of an operation instruction based on the operation behavior log includes:

[0015] Historical execution data of operation instructions is obtained based on operation behavior logs. The historical execution data includes the total number of historical executions of operation instructions, the standardization degree corresponding to each execution, and the average standardization degree of operation instructions in the process.

[0016] Based on the standardization degree corresponding to each execution, determine the standard deviation of the standardization degree corresponding to the operation instruction;

[0017] The stability of the operation instruction is determined based on the normality of each execution, the total number of historical executions, the mean of normality, and the standard deviation of normality.

[0018] In one embodiment, determining the stability of an operation instruction based on the normalization degree corresponding to each execution, the total number of historical executions, the mean normalization degree, and the standard deviation of the normalization degree includes:

[0019] If the total number of historical executions is greater than or equal to a preset threshold, the stability of the operation instruction is determined based on the normality of each execution, the total number of historical executions, the mean of normality, and the standard deviation of normality.

[0020] If the total number of historical executions is less than a preset threshold, the stability of the operation instruction is determined based on the normality of each execution, the total number of historical executions, the preset mean normality, and the preset standard deviation of normality.

[0021] In one embodiment, the step of splitting the process according to stability to obtain the corresponding operation segments of the process includes:

[0022] A stability sequence is constructed based on the stability and the execution order of the process.

[0023] Perform first-order differencing on the stability sequence and take the absolute value to determine the location of the maximum difference;

[0024] Based on the location of the maximum difference, the process is broken down to obtain the corresponding operation segments.

[0025] In one embodiment, calculating the image weights corresponding to each operation segment includes:

[0026] Obtain the stability of each operation instruction within the operation segment, the historical stability of the corresponding operation instruction, and the proportion of the operation segment duration to the total process duration;

[0027] The overall specificity index of the operation segment is determined based on the absolute value of the difference between the stability of each operation instruction and the historical stability.

[0028] The overall specificity index is multiplied by the proportion to obtain the image weight corresponding to the operation segment.

[0029] In one embodiment, the step of filtering target operation segments based on profile weights and determining the target credibility score by combining the duration difference coefficient of the operation vectors corresponding to the target operation segments in the historical benchmark process with the similarity of the operation vectors includes:

[0030] The initial credibility score is obtained by multiplying the duration difference coefficient and the similarity of the operation vector.

[0031] The initial credibility score is normalized to obtain the target credibility score.

[0032] In one embodiment, the verification process of the operational behavior based on the target credibility score to obtain the verification result includes:

[0033] The verification result is determined based on the first preset threshold, the second preset threshold, and the target credibility score; the first preset threshold is greater than the second preset threshold.

[0034] In one embodiment, determining the verification result based on a first preset threshold, a second preset threshold, and a target credibility score includes:

[0035] If the target credibility score is greater than or equal to the first preset threshold, the verification result is considered successful.

[0036] If the target credibility score is greater than the second preset threshold and less than the first preset threshold, the verification result is enhanced verification.

[0037] If the target credibility score is less than or equal to the second preset threshold, the verification result is that the verification fails.

[0038] In one embodiment, obtaining the operation behavior log and operation vector based on the operation behavior sequence includes:

[0039] The sequence of operational behaviors is structured to obtain an operational behavior log containing personnel number, operation number, operation duration, timestamp, and process batch number;

[0040] Each operation instruction in the sequence of operation behaviors is mapped to a corresponding standardized code, and arranged in the order of operation execution to form an operation vector with the same dimension as the number of operation instructions.

[0041] Secondly, another embodiment of the present invention provides a standardized operation intelligent verification and approval system based on dual identity authentication, the system comprising:

[0042] The parameter acquisition module is used to obtain the operation behavior sequence by recognizing personnel action data and equipment status change data, and based on the operation behavior sequence, obtain the operation behavior log and operation vector; based on the operation behavior log, it determines the standardization degree and stability degree of the operation instruction; the standardization degree is used to characterize the degree of compliance deviation of a single operation instruction; the stability degree is used to characterize the long-term stable execution capability of the same operation instruction under different execution environments.

[0043] The splitting module is used to split the process according to stability, obtain the corresponding operation segments of the process, and calculate the profile weight of each operation segment; the profile weight is used to characterize the representativeness of the operation segment to the specific operation characteristics of the personnel.

[0044] The verification module is used to filter target operation segments based on profile weights, combine the duration difference coefficient of the operation vectors of the target operation segments corresponding to the historical benchmark process with the similarity of the operation vectors, determine the target credibility score, and perform verification processing on the operation behavior based on the target credibility score to obtain the verification result.

[0045] Thirdly, in another embodiment of the present invention, an electronic device is provided, including a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the method described in the first aspect.

[0046] Fourthly, in another embodiment of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0047] The present invention has the following beneficial effects:

[0048] This invention obtains operation behavior sequences and corresponding operation behavior logs and operation vectors by integrating personnel action data and equipment status change data. Based on the operation behavior logs, it quantifies the standardization and stability of operation instructions. Combining stability with process decomposition, it obtains operation segments and calculates profile weights representing the representativeness of personnel-specific operation characteristics. Finally, it focuses on the target operation segment, determining the target credibility score and outputting the verification result through the similarity and duration difference coefficient of operation vectors of corresponding operation segments in adjacent processes. This not only effectively avoids the hidden risk of traditional single-operation-point verification failing to identify compliant single-operation-points but abnormal operation segments, and prevents process deviation risks caused by identity theft, but also strengthens the deep binding between identity and operation behavior, it achieves automated hierarchical approval of operation behavior verification. While ensuring the compliance and accuracy of process operations, it significantly reduces the cost of manual intervention and improves the efficiency of intelligent verification and approval of standardized operations. Attached Figure Description

[0049] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0050] Figure 1 This is a schematic flowchart illustrating a standardized intelligent verification and approval method for operations based on dual identity authentication, provided as an embodiment of the present invention.

[0051] Figure 2 This is a schematic diagram of the structure of a standardized intelligent verification and approval system for operations based on dual identity authentication, provided in one embodiment of the present invention.

[0052] Figure 3 This is a schematic diagram of the structure of an electronic device provided in one embodiment of the present invention. Detailed Implementation

[0053] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a standardized intelligent verification and approval method for dual-identity authentication based on the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0054] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0055] The following description, in conjunction with the accompanying drawings, details a specific scheme for a standardized intelligent verification and approval method for operations based on dual identity authentication provided by this invention.

[0056] This invention proposes a standardized intelligent verification and approval method for operations based on dual identity authentication. Please refer to [link / reference]. Figure 1 The diagram illustrates a schematic flowchart of a standardized intelligent verification and approval method for operations based on dual identity authentication, according to an embodiment of the present invention. The method includes the following steps:

[0057] Step S1: By identifying personnel action data and equipment status change data, an operation behavior sequence is obtained, and based on the operation behavior sequence, operation behavior logs and operation vectors are acquired.

[0058] Personnel motion data refers to quantifiable data reflecting the physical behavior of workers, collected through environmental sensing devices such as high-precision industrial cameras. For example, it may include limb operation trajectories, valve turning actions, tool holding and use actions, workstation movement trajectories, and interactive actions on the operating terminal interface.

[0059] Equipment status change data refers to status change data captured by environmental sensing devices, caused by human operation or the operation of the equipment itself. For example, it may include the number of valve rotations and angles, equipment start-up and shutdown status, tool usage status, changes in equipment operating parameters (pressure / liquid level, etc.), and switching status of the operating interface.

[0060] The operational behavior sequence is generated by a pre-trained You Only Look Once Object Detection Model (YOLO) that fuses and identifies personnel movement data and equipment status change data. Industrial cameras capture video frames, and the YOLO model simultaneously detects personnel actions (by detecting the human body using the YOLO model and extracting key points (joints, hand positions) using pose estimation models such as RTMPose and HRNet)) and equipment status changes. Effective operations are ordered according to their execution time, outputting a standardized set of operational instructions, such as disinfection, valve turning, and data recording, forming the operational behavior sequence.

[0061] The YOLO model takes video frames as input and detects two types of targets simultaneously: human motion targets (limbs, postures, operational behaviors) and equipment / component targets (valves, instruments, switches, buttons, indicator lights, displays, etc.).

[0062] YOLO classifies and determines the status of devices by using the YOLO model to identify the current status of devices based on the features of pre-trained device samples. Examples of such statuses include: valves (open, closed, rotating, opening degree), instruments (pointer position, numerical range), switches (on, off), indicator lights (on, off, color change), and displays (interface changes, parameter refresh).

[0063] The process of YOLO outputting device state changes includes the YOLO model recording the changes of the device from one state to the next in a time series. For example, a valve changes from 50% opening to 0% (completely closed), an instrument changes from 0.8MPa to 1.0MPa, and a switch changes from OFF to ON. This process of change is the change of device state.

[0064] The process of timing-binding and causal verification of personnel actions includes: YOLO aligns personnel actions with equipment state changes on the timeline. If the equipment state changes accordingly after a personnel action occurs, it is determined to be a valid operation; if there is only a personnel action and no equipment state change, it is determined to be an invalid action / empty operation, thereby ensuring that the operation is real and valid.

[0065] The standardized output results refer to the final output of the model, including: personnel action labels (e.g., turning a valve); equipment status change labels (e.g., valve closed); action start and end time, duration, and the system generates structured log fields such as operation number and operation duration based on these.

[0066] Furthermore, the operation logs are structured to generate a biometric identity (ID) that includes: the personnel number associated with the most recently verified biometric identifier; the operation number mapped from the action tag; the operation duration calculated from the time difference between the start and end frames of the action; a timestamp recording the operation completion time; and a process batch number serving as a unique identifier for this job process. This log provides a unified and reliable data foundation for subsequent calculations of standardization, stability, operation segmentation, profile weighting, and credibility scores.

[0067] An operation vector is a mathematical vector formed by mapping each operation instruction in the operation sequence to a standardized code (such as a digital code or a one-hot encoding) and arranging them in the execution order. Its dimension is the same as the number of operation instructions, and it is used to quantify the similarity of operation segments in subsequent adjacent processes (historical baseline process and current process).

[0068] For example, by deploying high-precision industrial cameras and other environmental sensing devices at the work site, two types of data are collected in real time: personnel action data (limb trajectory, operating posture, etc.) and equipment status change data (valve opening degree (which can be determined by detecting the position of visual feature points (valve stem or scale pointer) of valve handwheel or dial using YOLO model), number of turns / angle (which can be determined by detecting the position of key feature points (such as handle, reflective sticker) of valve handwheel using YOLO model and calculating its cumulative rotation angle / angle relative to the valve center), and changes in instrument parameters (which can be determined by detecting interactive elements (buttons, sliders, menu items) on the screen / panel using YOLO model and determining their status (pressed / released, selected / unselected)). The personnel action data and equipment status change data are then input into a pre-trained YOLO object detection model in the form of continuous video frames.

[0069] Furthermore, the YOLO model performs two-dimensional synchronous recognition on video frames, and finally outputs a standardized sequence of operational behaviors arranged in chronological order. The recognition process includes: detecting personnel and equipment targets; identifying personnel action types; identifying equipment state changes; action-state causal binding; and sorting by execution time to generate the sequence of operational behaviors.

[0070] The process of detecting personnel and equipment targets includes locating personnel, operating equipment, tools, valves, instruments, and other targets in the image.

[0071] The process of identifying the type of personnel action includes identifying action tags (e.g., disinfection, turning valves, recording data, turning on equipment, turning off valves, etc.).

[0072] The process of identifying changes in equipment status includes identifying the equipment status caused by actions (e.g., valves changing from open to closed; changes in instrument parameters; equipment changing from standby to operation, etc.).

[0073] The process of action-state causal binding involves YOLO matching and associating human actions with changes in device state in a timely manner, excluding invalid actions (such as fake operations or no device feedback), and retaining only real and valid operations.

[0074] The process of generating an operation behavior sequence by sorting by execution time includes: the model outputs a standardized set of operation instructions according to the chronological order of the actions, i.e.: operation behavior sequence = [disinfection, valve turning, data recording, ...].

[0075] The intelligent verification and approval system for standardized operations based on dual identity authentication processes the sequence of operational behaviors output by YOLO in a structured manner, automatically filling in 5 core fields to form a standard log that can be used for standardization and stability calculations.

[0076] Generate the following fields for each process: person_id, operation_code, operation duration, timestamp, and process_id.

[0077] The personnel ID (person_id) is taken from the identity ID (face / fingerprint) most recently verified through biometrics before the start of the operation, and serves as the identity anchor point for the entire process (e.g., an equipment inspection process; a valve opening and closing operation process; a material delivery process).

[0078] The operation number (operation_code) is a pre-defined standard operation instruction code mapped to the action tags (disinfection, valve turning, data recording, etc.) output by the intelligent verification and approval system based on dual authentication. For example: disinfection → OP001; valve turning → OP002; data recording → OP003.

[0079] The operation duration is calculated by YOLO from the time difference between the start frame and the end frame of the action, accurate to milliseconds / second.

[0080] The timestamp records the system standard time when the operation was completed.

[0081] The process batch number process_id refers to the unique identifier automatically assigned by the system to this complete job process (J), which is used to group all operations of the same job into one group.

[0082] For example, the standard log is shown in Table 1:

[0083] Table 1

[0084]

[0085] It should be noted that, in the embodiments of the present invention, the standardized operation behavior sequence is obtained by using dual data of personnel actions and equipment status changes as inputs and by fusing and recognizing them through the YOLO model. After being processed in a structured manner by the system, operation behavior logs for quantitative index calculation and operation vectors for similarity comparison are generated simultaneously, providing data for dynamic process analysis in subsequent dual identity authentication.

[0086] Furthermore, the step of obtaining the operation behavior log and operation vector based on the operation behavior sequence includes:

[0087] The sequence of operations is structured to obtain an operation log containing personnel number, operation number, operation duration, timestamp, and process batch number.

[0088] Among them, the personnel number refers to the unique identity identifier of the associated operator based on the biometric features (such as facial recognition or fingerprint) most recently verified. It is used to clarify the subject to which the operation belongs and is the carrier of static identity anchoring in dual identity authentication.

[0089] The operation number refers to the unique code assigned to each standard operation instruction in the work process (e.g., the specific number corresponding to closing valve A), which is used to identify the specific type of operation behavior and is a key related field for subsequent standardization and stability calculations.

[0090] Operation duration refers to the time it takes for an operator to complete a single operation instruction, which is directly taken from the duration of the action recorded by the environmental sensing device.

[0091] A timestamp is a record of the time when an operation instruction is completed. It is used to clarify the temporal sequence of operations and to provide a time reference for constructing stability sequences and dividing time windows.

[0092] The process batch number is a unique identifier for each complete work process, used to associate all operation instructions under the same process, ensuring that analyses such as standardization calculations and operation segment breakdowns are all carried out around the same work cycle.

[0093] Operation behavior logs refer to a standardized data set formed after structured processing, which includes five fields: personnel number, operation number, operation duration, timestamp, and process batch number. It is the data source for subsequent calculations of standardization, stability, profile weight, and credibility scores.

[0094] It should be noted that, in the embodiments of the present invention, by standardizing the operation behavior sequence of the operators, the unstructured raw operation data is transformed into a structured log containing identity identifier, operation type, execution duration, time base, and process attribution, providing unified data support for subsequent calculations.

[0095] Each operation instruction in the sequence of operation behaviors is mapped to a corresponding standardized code, and arranged in the order of operation execution to form an operation vector with the same dimension as the number of operation instructions.

[0096] Among them, the operation instructions refer to the specific work steps contained in the sequence of operation behaviors. They are the basic units that constitute the process (e.g., closing valve A; recording pressure value; starting the pump). Each instruction corresponds to a unique standard operation code.

[0097] Standardized coding refers to a unified format identifier that maps each operation instruction to a specific format for quantitative calculation of operation instructions. This includes forms such as numerical coding or one-hot coding, ensuring that the instructions can be used for similarity comparison and other algorithm calculations.

[0098] An operation vector is a mathematical vector formed by replacing each operation instruction in an operation sequence with its corresponding standardized code and arranging them in the original operation execution order. Its dimension is exactly the same as the number of operation instructions in the operation sequence, and it is used for quantitative calculation of the similarity of subsequent adjacent process operation segments.

[0099] It should be noted that, in the embodiments of the present invention, based on a standardized sequence of operation behaviors, each operation instruction in the sequence is mapped to a standardized code in a unified format and then arranged according to the original operation execution order, ultimately forming an operation vector whose dimension matches the number of operation instructions. This vector can be used to subsequently determine the consistency of operation behaviors through similarity calculation.

[0100] Step S2: Based on the operation behavior log, determine the standardization and stability of the operation instruction. The standardization is used to characterize the degree of compliance deviation of a single operation instruction. The stability is used to characterize the long-term stable execution capability of the same operation instruction under different execution environments.

[0101] Among them, the standardization degree refers to a quantitative indicator calculated based on operation behavior log data, which characterizes the degree of compliance deviation of a single operation instruction. It is calculated by combining the percentage of times the current personnel execute the instruction with the ratio of the instruction's execution time to the average execution time of the same instruction within the same process. The closer the value is to the reasonable range, the more the single operation complies with the process compliance requirements.

[0102] For example, the standardization of personnel p executing instruction j in process J can be expressed as:

[0103]

[0104] in, This represents the number of times that person p should theoretically execute instruction j in process J. This represents the total number of instructions that person p needs to execute in process J. This indicates the execution time of the current instruction j. This represents the average execution time of all instructions j in process J.

[0105] Stability refers to a quantitative indicator calculated based on accumulated data from historical operation logs, representing the long-term stable execution capability of the same operation instruction under different execution environments. It is quantified by the deviation of the standardization of each execution from the mean within the corresponding process and the overall standard deviation. When the number of historical executions is insufficient, a system preset value is used. The more stable the value, the stronger the consistency and anti-interference ability of the instruction execution.

[0106] The degree of compliance deviation refers to the deviation between the actual execution of a single operation instruction (such as the number of executions and the duration) and the standard execution of the same instruction within the same process. It is the basis for judging whether a single operation is compliant.

[0107] Different execution environments refer to the execution conditions of the same operation instruction under different processes, times, and scenarios (such as different batches of work, different personnel cooperation, and different equipment status), reflecting the differences in external variables of instruction execution.

[0108] Long-term stable execution capability refers to the level of consistency in compliance when operating instructions are executed multiple times in different environments. It is an indicator for judging the stability of personnel's operating habits and avoiding the risk of faking single operation points.

[0109] It should be noted that, in the embodiments of the present invention, operation behavior logs are used as data support to calculate the standardization degree, which represents the degree of compliance deviation of a single operation, and the stability degree, which represents the long-term stable execution capability of instructions, respectively, providing quantitative indicators for subsequent operation segment splitting, profile weight calculation and credibility determination.

[0110] It should be noted that the personnel ID in the operation behavior log is used to identify abnormal identities and temporarily lock accounts; the operation number and process batch number are used to mark abnormal operation segments; and the timestamp is used to trace the time when the operation occurred, providing accurate evidence for alarm triggering and manual review.

[0111] For example, based on the complete information recorded in the operation behavior log, when an abnormal process occurs, the entire operation process can be quickly traced using information such as personnel number, process batch number, and timestamp, providing a basis for safety investigation and responsibility determination.

[0112] Furthermore, the operation behavior log includes personnel number, operation number, operation duration, timestamp, and process batch number. The determination of the standardization level corresponding to the operation instruction based on the operation behavior log includes:

[0113] Extract the average duration, operation duration, and percentage of execution times from the operation behavior log.

[0114] The average duration refers to the average time taken for a specific operation instruction (identified by the operation number) to be completed by all executors within the same work process (identified by the process batch number). It is calculated by statistically analyzing all operation duration fields of the operation instruction in the operation behavior log and serves as a benchmark value for measuring the routine execution time of the operation instruction.

[0115] For example, operation time refers to the specific time spent by a single operator to complete a specific operation instruction. It is extracted directly from the duration field of the operation behavior log and is data reflecting the efficiency and compliance of the operator's single operation execution.

[0116] The percentage of execution times refers to the ratio of the number of times a person executes a specific operation instruction in the same work process to the total number of times that operation instruction is executed in that process. The higher the percentage, the greater the weight of the instruction in terms of the person's compliance with the process.

[0117] It should be noted that, in the embodiments of the present invention, three types of key parameters are extracted from the operation behavior log: the average duration of the normal execution of the instruction, the operation duration of the actual rhythm of a single operation, and the percentage of execution times of the instruction that reflects the weight of the instruction's impact on process compliance. This provides data support for subsequently quantifying the degree of compliance deviation (standardization) of a single operation instruction through formulas.

[0118] The standardization of operation instructions is determined based on the average duration, operation duration, and percentage of executions.

[0119] Among them, the operation instruction refers to a single operation task (such as turning a valve; recording data) defined according to a preset standard in the work process. Each instruction corresponds to a unique operation number and is the calculation unit of standardization. Its execution status is directly related to the compliance of the process.

[0120] Standardization refers to a quantitative indicator calculated based on the percentage of executions and the ratio of operation time to the average time. It is used to characterize the degree of compliance deviation of a single operation instruction. The closer the value is to the reasonable range, the more the single operation conforms to the unified execution standard within the process.

[0121] It should be noted that, in the embodiments of the present invention, the average duration of operation instructions within the same process is used as a benchmark. Combined with the specific operation time and execution frequency ratio of a single person executing the instruction (reflecting the influence weight of the instruction), the standardization degree characterizing the compliance deviation of a single operation is obtained through quantitative calculation, providing a quantitative basis for subsequent stability calculation, operation segment splitting and credibility determination.

[0122] Furthermore, determining the stability corresponding to the operation instruction based on the operation behavior log includes:

[0123] Historical execution data of operation instructions is obtained based on operation behavior logs. The historical execution data includes the total number of historical executions of operation instructions, the standardization degree corresponding to each execution, and the average standardization degree of operation instructions in the process.

[0124] Historical execution data refers to the set of quantitative data related to the past execution of operation instructions, which is accumulated and stored based on operation behavior logs. It is used to calculate instruction stability and support subsequent operation segment splitting and reliability determination.

[0125] The total number of historical executions refers to the total number of times a specific operation instruction has been executed in all historical job processes. It is obtained by counting the total number of records corresponding to the operation instruction in the operation behavior log. If the number of historical executions is insufficient, the system preset value should be used instead.

[0126] The compliance level of each execution refers to a quantitative indicator calculated based on the percentage of times an operation instruction is executed, the operation duration, and the average duration of the instruction within the corresponding process in each historical execution process. It is used to characterize the degree of compliance deviation of a single execution.

[0127] The average standardization of an operation instruction within a given process refers to the average standardization of all execution records of that operation instruction within a specific work process (identified by the process batch number) to which a particular historical execution of that operation instruction belongs. It is used to measure the routine compliance level of that instruction in a specific process environment.

[0128] It should be noted that, in the embodiments of the present invention, three types of historical execution datasets are extracted from the operation behavior log, including the total number of historical executions of operation instructions, the standardization of each execution, and the average standardization within the corresponding process, to provide data for subsequent calculation of the stability of operation instructions.

[0129] Based on the standardization degree corresponding to each execution, determine the standard deviation of the standardization degree corresponding to the operation instruction.

[0130] The standard deviation of compliance refers to the dispersion of the compliance value of an operational instruction across all historical execution counts. It is calculated by statistically analyzing all historical values ​​and is used to quantify the range of fluctuations in the compliance deviation of the instruction under different execution scenarios.

[0131] It should be noted that, in the embodiments of the present invention, based on the standardization of each execution of the operation instruction in history, the standard deviation of the standardization of the instruction is calculated through statistical analysis to quantify the fluctuation of its compliance deviation, providing parameter support for subsequent calculation of the stability of the operation instruction.

[0132] The stability of the operation instruction is determined based on the normality of each execution, the total number of historical executions, the mean of normality, and the standard deviation of normality.

[0133] Among them, stability refers to the long-term stable execution capability of operation instructions under different execution environments. It is calculated by integrating the standardization of each execution, the total number of historical executions, the mean of standardization, and the standard deviation. It is a parameter for subsequent process decomposition and profile weight calculation.

[0134] For example, the stability of instruction j It can be represented as:

[0135]

[0136] in, This indicates the total number of instructions j executed. This indicates the standardization of instruction j in process J during its i-th execution. This represents the average normality of instruction j in process J across multiple executions. This represents the standard deviation of the normality of instruction j in process J when executed multiple times.

[0137] It should be noted that, in the embodiments of the present invention, the total number of executions of the current instruction j is traversed. (Using the execution time order of different processes as the traversal order), extract the normalization degree at the i-th execution number. The mean normality of instruction j in process J to which this execution belongs. The difference represents the changes in the execution status of the current instruction under different execution environments during execution. This difference is further compared with the standard deviation of the instruction's regularity across all executed iterations. The ratio represents the difference between the current execution standard and the local execution, and its significance in the overall execution, reflecting the changes in the execution status of the current instruction during the execution process.

[0138] It should be noted that, in the embodiments of the present invention, based on the historical execution data of the operation instructions, four types of parameters are integrated: the standardization degree corresponding to each execution, the total number of historical executions, the mean of the standardization degree, and the standard deviation of the standardization degree. The stability degree, which characterizes the long-term stable execution capability of the operation instructions, is obtained through quantitative calculation, providing data support for subsequent process decomposition, profile weight calculation, and final credibility score.

[0139] Furthermore, determining the stability of the operation instruction based on the normalization degree corresponding to each execution, the total number of historical executions, the mean normalization degree, and the standard deviation of the normalization degree includes:

[0140] If the total number of historical executions is greater than or equal to a preset threshold, the stability of the operation instruction is determined based on the normality of each execution, the total number of historical executions, the mean of normality, and the standard deviation of normality.

[0141] The preset threshold refers to the critical value set by the system to determine whether there is sufficient historical execution data (for example, the preset threshold can be 10 times). When the total number of historical executions reaches or exceeds this value, the stability is calculated using real historical data; if it does not reach this value, the system's preset standard mean and variance are used instead.

[0142] It should be noted that, in the embodiments of the present invention, based on real accumulated historical data, four types of parameters are integrated: the normalization degree corresponding to each execution, the total number of historical executions, the mean of normalization degree, and the standard deviation of normalization degree. The stability degree, which characterizes the long-term stable execution capability of the instruction, is calculated through a quantitative formula, providing data support for subsequent process decomposition, profile weight calculation, and final credibility score.

[0143] If the total number of historical executions is less than a preset threshold, the stability of the operation instruction is determined based on the normality of each execution, the total number of historical executions, the preset mean normality, and the preset standard deviation of normality.

[0144] The preset standardization mean refers to the average standardization value set in advance by the system. When the total number of historical executions of an operation command is insufficient, the average standardization value of the command in each historical process is used to replace it in the calculation to ensure the continuity and rationality of the stability calculation.

[0145] The preset standard deviation of normality refers to the standard normality dispersion index preset by the system. When the total number of historical executions of an operation command is insufficient, the standard deviation of the normality of the command under all execution counts is used to participate in the calculation to quantify the fluctuation range of normality.

[0146] It should be noted that, in the embodiments of the present invention, when the total number of historical executions of the operation instruction is less than a preset threshold (e.g., 10 times), and it is impossible to calculate the mean and standard deviation of the standardization degree using sufficient historical data, the system's preset mean and standard deviation of the standardization degree are used, combined with the existing standardization degree corresponding to each execution and the total number of historical executions, to complete the stability calculation, ensuring that even in the case of insufficient data, it can still provide a quantitative basis for subsequent operation segment splitting and intelligent verification and approval.

[0147] Step S3: Decompose the process according to stability to obtain the corresponding operation segments, and calculate the profile weight of each operation segment; the profile weight is used to characterize the representativeness of the operation segment to the specific operation characteristics of the personnel.

[0148] In this context, a process refers to a set of continuous operations executed in a pre-defined standard sequence around a specific operational objective (such as chemical production or equipment maintenance). It has a unique process batch number and is the overall object of the operation segment breakdown.

[0149] Operation segment splitting refers to arranging the operation instructions in the process according to their stability to form a stability sequence, performing first-order difference on the sequence and taking the absolute value, and using the Automatic Multiscale-based Peak Detection Algorithm (AMPD) to extract the position of the maximum difference with the most drastic change in stability, and breaking along this position to obtain multiple continuous intervals. Within each interval, the change in instruction stability is small and the operation scenario and behavior characteristics are consistent.

[0150] An operation segment refers to the key analysis unit formed after the process is broken down. It is a continuous interval with significantly consistent operational stability (e.g., valve turning operation segment; disinfection operation segment). Each operation segment corresponds to a specific work scenario and is the basic unit for calculating profile weights.

[0151] Profile weight refers to a parameter obtained by weighting the difference between the stability of instructions within the operation segment and the historical stability, and the proportion of the operation segment duration to the total process duration. It is used to quantify the representativeness of the operation segment to the specific operational characteristics of the personnel.

[0152] For example, the image weight of operation segment D can be expressed as:

[0153]

[0154] in, This indicates the normality of instruction j within the current operation segment. This represents the historical stability mean of instruction j. To prevent the use of small positive numbers with a denominator of zero (such as 0.001). Indicates the duration of the operation segment. Indicates the total duration of the process.

[0155] Personnel-specific operational characteristics refer to the personalized behavioral patterns (such as operating time habits, action execution rhythm, etc.) formed by operators when performing specific operations over a long period of time. These characteristics are stable and unique, and are the core features that distinguish different operators.

[0156] Representativeness refers to the degree to which an operation segment can accurately reflect the unique operational characteristics of an individual. The higher the profile weight value, the better the operation segment reflects the individual's personalized operating habits, and the greater its reference value for determining the match between identity and operational behavior.

[0157] It should be noted that, in the embodiments of the present invention, the complete work process is divided based on stability to obtain multiple operation segments with consistent operation stability; then, by combining the specificity (stability difference) and time proportion of each operation segment, the profile weight that characterizes the representativeness of the operation segment to the specific operation characteristics of the personnel can be calculated.

[0158] Furthermore, the process is further divided based on stability to obtain the various operation segments corresponding to the process, including:

[0159] A stability sequence is constructed based on the stability and the execution order of the process.

[0160] Among them, the process execution order refers to the sequential execution logic of all operation instructions in the work process according to the preset standard (such as disinfection, valve turning, and data recording), which is consistent with the order of the timestamps in the operation behavior log, and provides a basis for the arrangement of the stability sequence.

[0161] A stability sequence refers to an ordered set of data containing the stability of each instruction (e.g., stability of disinfection instruction, stability of turning instruction, stability of recording instruction) formed by arranging all operation instructions within a single complete work process (identified by the process batch number) in the order of their execution. This provides a basis for the subsequent breakdown of operation segments.

[0162] It should be noted that, in the embodiments of the present invention, based on the stability of each operation instruction, the stability of all instructions is arranged sequentially according to the preset execution order of the operation instructions in the work process (consistent with the actual execution order) to form a structured stability sequence, providing orderly data support for subsequent extraction of operation segment splitting points and accurate process splitting through first-order difference and AMPD algorithms.

[0163] Perform first-order differencing on the stability sequence and take the absolute value to determine the location of the maximum difference.

[0164] Among them, the stability sequence refers to the ordered data set formed by arranging the stability of each operation instruction in the order of execution of the operation instructions in the work process (for example, the stability sequence for process disinfection, valve turning and recording data is (w1, w2, w3)), which is the analysis object of process decomposition.

[0165] It should be noted that different types of operation instructions have different long-term stability benchmark values, and the stability sequence is used to reflect the switching of task phases.

[0166] For example, the first-order difference refers to the calculation of subtracting the preceding term from the following term for two consecutive stability values ​​in a stability sequence (e.g., the first-order difference of the sequence (w1, w2, w3) is (w2-w1, w3-w2)), used to characterize the change in stability of adjacent instructions.

[0167] For example, taking the absolute value means performing an absolute value transformation on the difference result obtained from the first-order difference calculation, eliminating the influence of positive and negative signs, and only retaining the magnitude of the change (e.g., when the difference is -0.3, the absolute value becomes 0.3).

[0168] The maximum difference position refers to the location in the original stability sequence corresponding to the point with the largest value in the difference sequence formed after first-order differencing and taking the absolute value. This position corresponds to a node where operational stability changes significantly (such as switching from a disinfection step to a valve operation step) and serves as the basis for process decomposition.

[0169] It should be noted that, in the embodiments of the present invention, according to the stability sequence arranged in the order of operations, the change range of stability of adjacent instructions is calculated by first-order difference, and after absolute value conversion, the position of the maximum difference with the most drastic change is selected, which provides the basis for splitting nodes for subsequent process splitting along the position to obtain operation segments with consistent operation stability.

[0170] Based on the location of the maximum difference, the process is broken down to obtain the corresponding operation segments.

[0171] It should be noted that, in the embodiments of the present invention, the maximum difference position (operation stability mutation node) obtained by stability sequence analysis is used as the split boundary to divide the complete operation process into multiple operation segments with consistent operation scenarios and behavioral characteristics, providing a structured analysis unit for subsequent calculation of the profile weight and credibility determination of each operation segment.

[0172] Furthermore, the calculation of the image weights corresponding to each operation segment includes:

[0173] Obtain the stability of each operation instruction within the operation segment, the historical stability of the corresponding operation instruction, and the proportion of the operation segment duration to the total process duration.

[0174] The stability of each operation instruction within the operation segment is a quantitative indicator calculated based on the accumulated data of historical operation behavior logs. It is used to characterize the long-term stable execution capability of the instruction under different execution environments and is quantified by the deviation of the standardization of each execution from the mean of the corresponding process and the overall standard deviation.

[0175] Historical stability refers to the stability data accumulated during all past executions of an operation instruction, including the stability at the last execution of the instruction (as a standard stability benchmark) and the historical stability mean (the average stability of the instruction across all executed processes). It is used to compare and analyze the specificity of instruction stability within the current operation segment.

[0176] Operation segment duration refers to the total time consumed by a single operation segment, which is the sum of the operation times of all operation instructions within the segment. It is directly extracted from the operation behavior log.

[0177] The total process duration refers to the total time consumed by the complete work process (identified by the process batch number), which is the sum of the operation times of all operation instructions within the process. It is the benchmark for calculating the proportion of operation time.

[0178] The proportion of operation segment time to total process time refers to the ratio of operation segment time to total process time. It is used to quantify the time weight of operation segment in the whole process. The higher the time proportion, the easier it is to strengthen the representativeness of the operation segment to the characteristics of personnel operation.

[0179] It should be noted that, in the embodiments of the present invention, the current stability of each operation instruction within the operation segment, the historical stability (benchmark and mean) of each operation instruction, and the proportion of the operation segment duration to the total process duration need to be extracted from the operation behavior log and historical data, so as to provide data support for subsequent calculation of profile weights through specific quantification and time proportion weighting.

[0180] The overall specificity index of the operation segment is determined based on the absolute value of the difference between the stability of each operation instruction and the historical stability.

[0181] The absolute value of the difference refers to the result of taking the absolute value of the difference between the stability of the operation instruction in the current operation segment and the historical stability. It is used to eliminate the influence of positive and negative signs and only retain the deviation between the two.

[0182] The overall specificity index refers to the value obtained by normalizing and summing the absolute values ​​of the differences between the stability of all operation instructions within the operation segment and the historical stability. It is used to quantify the degree to which the operation segment deviates from the normal execution characteristics of the instructions and is an intermediate parameter for the calculation of profile weight.

[0183] It should be noted that, in the embodiments of the present invention, based on each operation instruction within the operation segment, the absolute value of the difference between the current stability and the historical stability of each instruction is calculated, and integrated to obtain a specific index characterizing the overall deviation of the operation segment from the normal execution characteristics, providing data for subsequent calculation of profile weights in conjunction with time proportions.

[0184] The overall specificity index is multiplied by the proportion to obtain the image weight corresponding to the operation segment.

[0185] The overall specificity index refers to the quantitative score obtained by summing the specific contributions of all operation instructions within an operation segment. It reflects the overall difference between the execution characteristics of multiple instructions within the operation segment and their historical stability. Its calculation is based on the absolute value of the difference between the current stability and historical stability of each instruction within the segment, divided by the mean stability of that instruction (adding 0.001 to avoid a denominator of 0), and then summing the results of all instructions. A higher score indicates that the operation segment better reflects the specific operating habits of the personnel.

[0186] For example, the overall specificity index of operation segment D It can be represented as:

[0187]

[0188] in, This indicates the stability of instruction j in operation segment D. This indicates the last time instruction j was executed. stability, This represents the mean stability of instruction j. To prevent the use of small positive numbers with a denominator of zero (such as 0.001).

[0189] It should be noted that, in the embodiments of the present invention, the total number of operations in operation segment D is first calculated. Stability of instruction j at position i , and instruction j at the last moment of execution stability The absolute value of the difference, and the mean stability of instruction j (including only the current instruction type j). Compared to the previous one, this method shows the special execution situation of the instruction j in the segment, making the instructions contained in the current segment more specific and more representative of the operation of the current person.

[0190] The ratio of the operation segment duration to the total duration of the corresponding complete work process is used to quantify the time weight of the operation segment in the entire process. The higher the time weight, the easier it is to strengthen the representativeness of the operation segment to the characteristics of personnel operation.

[0191] The profile weight can be calculated by multiplying the overall specificity index by the proportion of the operation segment duration to the total process duration. It is used to characterize the representativeness of the operation segment to the specific operation characteristics of the personnel. The higher the value, the greater the reference value of the operation segment in the credibility score. It is an indicator for constructing a profile of the identity and operation segment.

[0192] It should be noted that, in the embodiments of the present invention, the profile weight is obtained by multiplying the overall specificity index of all instructions within the operation segment and the duration ratio of the operation segment in the complete process (assigning time weight), thereby ultimately realizing the quantification of the operation segment's ability to represent the unique operational characteristics of the personnel, and providing data support for the subsequent determination of the credibility of the operation process.

[0193] Step S4: Filter target operation segments according to the profile weight, combine the duration difference coefficient of the operation vector of the target operation segment corresponding to the historical benchmark process and the similarity of the operation vector to determine the target credibility score, and perform operation behavior verification processing based on the target credibility score to obtain the verification result.

[0194] Among them, the target operation segment refers to the key operation segment (high profile weight operation segment) that is selected based on the profile weight and is highly representative of the operation characteristics of the personnel. It is the evaluation object of the subsequent credibility score.

[0195] Historical benchmark processes refer to two complete work processes that are executed consecutively in a preset order around the same work objective. Each process has a unique batch number, which is used to compare the consistency of operational behavior.

[0196] An operation vector is a mathematical vector formed by mapping each operation instruction in a sequence of operation behaviors to a standardized code (such as numeric code or one-hot code) and arranging them in the order of execution. Its dimension is the same as the number of operation instructions, and it is used for similarity quantification calculation.

[0197] The duration difference coefficient refers to a parameter that quantifies the degree of deviation in execution time between two adjacent process corresponding to the target operation segment. The closer the coefficient is to 1, the smaller the difference in execution time between the two operation segments and the stronger the consistency.

[0198] Operation vector similarity refers to an indicator that measures the degree of matching between the operation vectors of corresponding target operation segments in two adjacent processes. It is calculated by normalized score of edit distance or cosine similarity after one-hot encoding. The higher the similarity, the more consistent the operation behavior pattern.

[0199] The target credibility score refers to the final score calculated based on the duration difference coefficient of the target operation segment and the similarity of the operation vector, and normalized to the [0, 1] interval by the S-shaped function (Sigmoid Function). It is used to quantify the credibility of the match between the operation behavior and the personnel identity.

[0200] For example, verification processing refers to the graded verification actions performed on operational behaviors based on the target credibility score, including three types of processing methods: automatic verification passed, triggered enhanced verification, and verification failed.

[0201] The verification result refers to the final judgment output after verification processing, which may include: verification passed, enhanced verification, and verification failed, directly guiding the approval decision of the work process.

[0202] It should be noted that, in the embodiments of the present invention, the image weight is used as the screening criterion to focus on the target operation segment with high representativeness. By integrating the similarity of operation vectors and the duration difference coefficient of the target operation segments corresponding to adjacent processes, the target credibility score with quantitative credibility is calculated. Then, based on the score, a hierarchical verification process is performed, and finally the verification result guiding the approval process is output.

[0203] Furthermore, the step of filtering target operation segments based on profile weights, and determining the target credibility score by combining the duration difference coefficient of the operation vectors corresponding to the target operation segments in the historical benchmark process with the similarity of the operation vectors, includes:

[0204] The initial credibility score is obtained by multiplying the duration difference coefficient and the similarity of the operation vector.

[0205] The initial credibility score refers to the raw score calculated by multiplying the duration difference coefficient and the similarity of the operation vector. It has not been normalized and is used to initially quantify the compliance consistency between the current operation segment and historical operation segments of the same type. It is the basis for obtaining the final credibility score by normalization using the Sigmoid function.

[0206] It should be noted that, in the embodiments of the present invention, the same type of operation segments in the preceding and following processes are taken as the analysis objects. The initial credibility score is obtained by combining the time difference coefficient of the two operations with the similarity of the operation vectors, thereby achieving a preliminary quantification of the compliance consistency of the operation segments.

[0207] The initial credibility score is normalized to obtain the target credibility score.

[0208] Normalization refers to the standardization operation of mapping the initial confidence score to the [0, 1] interval using the Sigmoid function, eliminating the problem of inconsistent original score ranges and making the scores more comparable and practical for judgment.

[0209] The target credibility score refers to the final credibility quantification result obtained after normalization, with the value range limited to [0, 1], providing a basis for judgment in subsequent hierarchical verification processing.

[0210] For example, the target credibility score of the current operation segment It can be represented as:

[0211]

[0212] in, This represents the historical baseline process of the current operation and the operation vector under the current process. Similarity, This represents the difference in execution time between the operation vectors corresponding to the two process segments. Indicates the coefficient of variation in duration Constraints are in place to prevent negative scores.

[0213] It should be noted that, in the embodiments of the present invention, the initial credibility score calculated based on the similarity of operation vectors and the difference coefficient of duration is normalized by the Sigmoid function to standardize the score to the [0, 1] interval, thereby obtaining a target credibility score with a unified judgment standard, which provides a quantitative basis for subsequent hierarchical approval decisions.

[0214] Furthermore, the verification process based on the target credibility score to obtain the verification result includes:

[0215] The verification result is determined based on the first preset threshold, the second preset threshold, and the target credibility score; the first preset threshold is greater than the second preset threshold.

[0216] The first preset threshold refers to the high credibility judgment threshold set in advance by the system (for example, the first preset threshold can be 0.75), which is the standard for distinguishing between verification passing and enhanced verification. The value of the first preset threshold is higher than the second preset threshold.

[0217] The second preset threshold refers to the low confidence threshold value set in advance by the system (for example, the second preset threshold can be 0.35), which is the standard for distinguishing between enhanced verification and verification failure.

[0218] The target credibility score refers to the final credibility quantification result obtained after normalization by the Sigmoid function. The value range is limited to the interval [0, 1]. It is calculated by the similarity of operation vectors and the difference in duration between adjacent target operation segments and is the core basis for judging the verification results.

[0219] The verification result refers to the graded judgment conclusion based on the comparison between the target credibility score and two preset thresholds, including three types of results: verification passed, enhanced verification, and verification failed.

[0220] It should be noted that, in the embodiments of the present invention, two preset threshold values ​​are used as the judgment criteria to classify the target credibility score ([0,1] interval) into the corresponding score interval, thereby outputting the verification result of verification passed, enhanced verification, or verification failed, providing judgment rules for realizing hierarchical intelligent approval of work processes.

[0221] Further, determining the verification result based on the first preset threshold, the second preset threshold, and the target credibility score includes:

[0222] If the target credibility score is greater than or equal to the first preset threshold, the verification result is "verification passed".

[0223] Among them, "verification passed" refers to the verification result when the target credibility score reaches or exceeds the first preset threshold, indicating that the operation behavior is highly consistent with the personnel's exclusive operation characteristics, and the process can be approved to continue without additional verification.

[0224] If the target credibility score is greater than the second preset threshold but less than the first preset threshold, the verification result is enhanced verification.

[0225] Among them, enhanced verification refers to the verification result when the target credibility score is between the second preset threshold and the first preset threshold, indicating that there is a certain degree of uncertainty in the operation behavior, and further confirmation is required through additional means such as secondary biometric verification and manual review.

[0226] If the target credibility score is less than or equal to the second preset threshold, the verification result is that the verification fails.

[0227] Among them, verification failure refers to the verification result when the target credibility score is lower than or equal to the second preset threshold. This indicates that the operation behavior is significantly different from the personnel's exclusive operation characteristics, and there are risks such as identity theft and process abnormality. The process should be interrupted immediately and an alarm should be triggered.

[0228] It should be noted that, in the embodiments of the present invention, the first preset threshold (high credibility threshold) and the second preset threshold (low credibility threshold) are used as the dividing criteria to classify the target credibility score in the [0, 1] interval into three intervals, and the verification results of verification passed, enhanced verification, and verification failed can be output respectively, so as to realize the accurate classification of the credibility of the operation behavior and provide an automated and differentiated approval decision basis for the work process.

[0229] In summary, this invention obtains operational behavior sequences and corresponding operational behavior logs and operational vectors by integrating personnel action data and equipment status change data. Based on the operational behavior logs, it quantifies the standardization and stability of operational instructions. Combining stability with process decomposition, it obtains operational segments and calculates profile weights representing their representativeness to personnel-specific operational characteristics. Finally, it focuses on the target operational segment, determining the target credibility score and outputting the verification result through the similarity and duration difference coefficient of operational vectors of corresponding operational segments in adjacent processes. This not only effectively avoids the hidden risks of traditional single-operation-point verification failing to identify compliant single-operation-points but abnormal operational segments, and prevents process deviation risks caused by identity theft, but also strengthens the deep binding between identity and operational behavior, it also achieves automated hierarchical approval of operational behavior verification. While ensuring the compliance and accuracy of process operations, it significantly reduces the cost of manual intervention and improves the efficiency of intelligent verification and approval of standardized operations.

[0230] This invention proposes a standardized intelligent verification and approval system for operations based on dual identity authentication. Please refer to [link / reference]. Figure 2 The diagram illustrates a structural schematic of a standardized intelligent verification and approval system 200 based on dual identity authentication, according to an embodiment of the present invention. The system includes:

[0231] The parameter acquisition module 201 is used to obtain the operation behavior sequence by recognizing personnel action data and equipment status change data, and to obtain the operation behavior log and operation vector based on the operation behavior sequence; based on the operation behavior log, it determines the standardization degree and stability degree of the operation instruction; the standardization degree is used to characterize the degree of compliance deviation of a single operation instruction; the stability degree is used to characterize the long-term stable execution capability of the same operation instruction under different execution environments;

[0232] The splitting module 202 is used to split the process according to stability, obtain the operation segments corresponding to the process, and calculate the profile weight corresponding to each operation segment; the profile weight is used to characterize the representativeness of the operation segment to the specific operation characteristics of the personnel.

[0233] The verification module 203 is used to filter target operation segments according to the profile weight, combine the duration difference coefficient of the operation vector of the target operation segment corresponding to the historical benchmark process and the similarity of the operation vector to determine the target credibility score, and perform verification processing of the operation behavior according to the target credibility score to obtain the verification result.

[0234] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer device can be divided into different functional modules to complete all or part of the functions described above. In addition, the intelligent verification and approval system for standardized operations based on dual identity authentication and the intelligent verification and approval method for standardized operations based on dual identity authentication provided in the above embodiments belong to the same concept. The specific implementation process is detailed in the method embodiment and will not be repeated here.

[0235] This invention also provides an electronic device. Please refer to [link to relevant documentation]. Figure 3 The electronic device may include a processor 301, a memory 302, and a program 3021 stored in the memory 302 and capable of running on the processor 301.

[0236] When program 3021 is executed by processor 301, it can achieve the following: Figure 1 Any steps in the corresponding method embodiments and the achievement of the same beneficial effects will not be repeated here.

[0237] Those skilled in the art will understand that all or part of the steps of the methods described in the above embodiments can be implemented by hardware related to program instructions, and the program can be stored in a readable medium.

[0238] This invention also provides a readable storage medium storing a computer program, which, when executed by a processor, can perform the above-described functions. Figure 1 Any step in the corresponding method embodiment can achieve the same technical effect, and will not be repeated here to avoid repetition.

[0239] The computer-readable storage medium of this invention can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. For example, a computer-readable storage medium can be an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0240] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0241] The program code contained on the storage medium can be transmitted using any suitable medium, including but not limited to wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0242] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming 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 terminal. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0243] This invention also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned steps to implement the intelligent verification and approval method for standardized operations based on dual authentication provided in the above embodiments.

[0244] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0245] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A standardized intelligent verification and approval method for operations based on dual identity authentication, characterized in that, The method includes: By identifying personnel action data and equipment status change data, an operation behavior sequence is obtained, and based on the operation behavior sequence, an operation behavior log and operation vector are acquired. Based on the operation behavior log, the standardization degree and stability degree corresponding to the operation instruction are determined; the standardization degree is used to characterize the degree of compliance deviation of a single operation instruction; the stability degree is used to characterize the long-term stable execution capability of the same operation instruction under different execution environments. The process is divided according to the stability to obtain the operation segments corresponding to the process, and the profile weight corresponding to each operation segment is calculated; the profile weight is used to characterize the representativeness of the operation segment to the specific operation characteristics of the personnel. The target operation segment is selected based on the image weight. The duration difference coefficient of the operation vector of the target operation segment corresponding to the historical benchmark process and the similarity of the operation vector are combined to determine the target credibility score. The operation behavior is then verified based on the target credibility score to obtain the verification result.

2. The standardized operation intelligent verification and approval method based on dual identity authentication according to claim 1, characterized in that, The operation behavior log includes personnel number, operation number, operation duration, timestamp, and process batch number. Determining the standardization of the operation instruction based on the operation behavior log includes: Extract the average duration, operation duration, and percentage of execution times from the operation behavior log; The standardization of the operation instruction is determined based on the average duration, the operation duration, and the percentage of execution times.

3. The standardized operation intelligent verification and approval method based on dual identity authentication according to claim 2, characterized in that, The step of determining the stability corresponding to the operation instruction based on the operation behavior log includes: Based on the operation behavior log, the historical execution data of the operation instruction is obtained. The historical execution data includes the total number of historical executions of the operation instruction, the standardization degree corresponding to each execution, and the average standardization degree of the operation instruction in the process. Based on the normality corresponding to each execution, determine the standard deviation of the normality corresponding to the operation instruction; The stability of the operation instruction is determined based on the normality corresponding to each execution, the total number of historical executions, the mean of the normality, and the standard deviation of the normality.

4. The standardized operation intelligent verification and approval method based on dual identity authentication according to claim 3, characterized in that, The step of determining the stability of the operation instruction based on the normalization degree corresponding to each execution, the total number of historical executions, the mean normalization degree, and the standard deviation of the normalization degree includes: If the total number of historical executions is greater than or equal to a preset threshold, the stability of the operation instruction is determined based on the normality corresponding to each execution, the total number of historical executions, the mean of the normality, and the standard deviation of the normality. If the total number of historical executions is less than the preset threshold, the stability of the operation instruction is determined based on the normality corresponding to each execution, the total number of historical executions, the preset normality mean, and the preset normality standard deviation.

5. The standardized operation intelligent verification and approval method based on dual identity authentication according to claim 1, characterized in that, The process is divided according to the stability to obtain the various operation segments corresponding to the process, including: Based on the stability, a stability sequence is constructed according to the process execution order; Perform a first-order difference on the stability sequence and take the absolute value to determine the location of the maximum difference; Based on the location of the maximum difference, the process is divided into various operation segments corresponding to the process.

6. The standardized operation intelligent verification and approval method based on dual identity authentication according to claim 1, characterized in that, The calculation of the image weight corresponding to each of the operation segments includes: The stability of each operation instruction within the operation segment, the historical stability of the operation instruction, and the proportion of the operation segment duration to the total process duration are obtained. The overall specificity index of the operation segment is determined based on the absolute value of the difference between the stability of each operation instruction and the historical stability. The overall specificity index is multiplied by the ratio to obtain the image weight corresponding to the operation segment.

7. The standardized operation intelligent verification and approval method based on dual identity authentication according to claim 1, characterized in that, The step of filtering target operation segments based on the portrait weights, and determining the target credibility score by combining the duration difference coefficient of the operation vectors of the target operation segments corresponding to the historical benchmark process and the similarity of the operation vectors, includes: An initial credibility score is obtained by multiplying the duration difference coefficient and the similarity of the operation vector. The initial credibility score is normalized to obtain the target credibility score.

8. The standardized operation intelligent verification and approval method based on dual identity authentication according to claim 1, characterized in that, The verification process based on the target credibility score to obtain the verification result includes: The verification result is determined based on the first preset threshold, the second preset threshold, and the target credibility score; the first preset threshold is greater than the second preset threshold.

9. The standardized operation intelligent verification and approval method based on dual identity authentication according to claim 8, characterized in that, The step of determining the verification result based on the first preset threshold, the second preset threshold, and the target credibility score includes: If the target credibility score is greater than or equal to the first preset threshold, the verification result is that the verification is passed. If the target credibility score is greater than the second preset threshold and less than the first preset threshold, the verification result is an enhanced verification. If the target credibility score is less than or equal to the second preset threshold, the verification result is that the verification fails.

10. The standardized operation intelligent verification and approval method based on dual identity authentication according to claim 2, characterized in that, The step of obtaining the operation behavior log and operation vector based on the operation behavior sequence includes: The sequence of operation behaviors is structured to obtain the operation behavior log containing the personnel number, the operation number, the operation duration, the timestamp, and the process batch number; Each operation instruction in the operation sequence is mapped to a corresponding standardized code, and arranged in the order of operation execution to form an operation vector with the same dimension as the number of operation instructions.

Images