Standard man-hour dynamic analysis method and system based on pure digital action code library

Abstract

The application provides a standard man-hour dynamic analysis method and system based on a pure digital action code library, and relates to the technical field of data processing.The method comprises the following steps: according to an initial action instruction sequence, performing matching in a preset pure digital action code library according to a bit sequence interval, obtaining a basic action feature set which is composed of an action major category interval corresponding to a first field and a subsequent field corresponding to a serial number of a detailed item; according to the serial number of the detailed item in the basic action feature set, searching for a corresponding action name and a basic time measurement value from a preset standard action parameter benchmark table, and collecting and forming a basic man-hour parameter matrix.The application performs dynamic working condition compensation on a basic TMU value, and improves the estimation accuracy of the standard man-hour in a complex operation scene and the adaptation ability of the actual production rhythm.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and system for dynamic parsing of standard working hours based on a pure digital action code library. Background Technology

[0002] In the field of digital management of standard working hours, especially in the discrete manufacturing industry of garments, it has become an industry practice to use pure digital codes to represent work actions. Most existing standard working hour systems have pre-established action code libraries, mapping various work actions, such as picking up and placing materials, machine operation, and sewing, to specific digital codes and associating them with preset time measurement units (TMU). In actual application, the system receives the digital code string input by the user, directly retrieves the corresponding standard time value through a table lookup method, and accumulates it to calculate the standard working hours of the process.

[0003] However, existing analytical methods, when processing continuous action sequences, mostly focus only on the independent time consumption of individual actions, potentially neglecting the dynamic characteristics of the action transition process. Specifically, during the operation, at the junction of two actions, the operator's limb movement state, such as changes in speed and acceleration, is not constant. For example, when an operator completes a large-displacement "picking and placing of cut pieces" action and then immediately performs a high-precision "point-to-point" action, the limbs need to transition from high-dynamic movement to a stationary state in a short time. This instantaneous speed change will generate a significant time impact, causing the actual operation time to be higher than the simple sum of the standard values ​​of individual actions. Some existing methods lack a quantitative evaluation mechanism for this "action transition time impact rate," and may not be able to dynamically correct the basic TMU value based on the spatial geometric characteristics of the actual action path, such as turning angles and displacement vector changes. This results in a systematic deviation between the calculated results and the actual operation time when calculating the time of processes involving complex action transitions, especially high-precision assembly or sewing processes with frequent start-stop operations, affecting the guidance of standard time data. Summary of the Invention

[0004] This invention provides a method and system for dynamic parsing of standard working hours based on a pure digital motion code library. It performs dynamic working condition compensation on the basic TMU value, improving the accuracy of standard working hours prediction and the adaptability to actual production rhythm in complex work scenarios.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] Firstly, a standard dynamic time parsing method based on a pure digital action code library, the method comprising:

[0007] Receive the pure numeric action code to be parsed, and convert the pure numeric action code to be parsed into an initial action instruction sequence according to the character position order;

[0008] Based on the initial action instruction sequence, the system matches the pre-set pure digital action code library according to the positional interval to obtain a basic action feature set composed of the action category interval corresponding to the first field and the detailed item sequence number corresponding to the subsequent fields.

[0009] Based on the detailed item sequence number in the basic motion feature set, the corresponding motion name and basic time measurement value are retrieved from the preset standard motion parameter benchmark table and summarized to form a basic working time parameter matrix.

[0010] Based on the basic working time parameter matrix, the instantaneous changes of adjacent basic time measurements at the action handover node are compared by difference to obtain the action handover time impact rate sequence.

[0011] Based on the spatial pose coordinates of each action handover node in the basic action feature set, the nodes are connected sequentially to form an action path polygon. The ratio of the minimum feature radius to the area of ​​the action path polygon is calculated. Based on the ratio, the action handover time impact rate sequence is scaled to obtain the corrected time impact rate sequence.

[0012] The corrected time impact rate sequence is subjected to time-series discrete accumulation operation to obtain the overall fatigue cumulative effect value; based on the overall fatigue cumulative effect value, the basic time measurement values ​​in the basic working time parameter matrix are corrected for the working conditions to obtain the target working time data.

[0013] Secondly, a standard time dynamic parsing system based on a pure digital action code library includes:

[0014] The parsing module is used to receive the pure numeric action code to be parsed and convert the pure numeric action code into an initial action instruction sequence according to the character position order;

[0015] The matching module is used to match the initial action instruction sequence in a preset pure digital action code library according to the positional interval, and obtain a basic action feature set composed of the action category interval corresponding to the first field and the detailed item sequence number corresponding to the subsequent fields.

[0016] The retrieval module is used to retrieve the corresponding action name and basic time measurement value from the preset standard action parameter benchmark table based on the detailed item sequence number in the basic action feature set, and summarize them to form a basic working time parameter matrix.

[0017] The calculation module is used to compare the instantaneous changes of adjacent basic time measurements at the action handover node based on the basic time parameter matrix to obtain the action handover time impact rate sequence.

[0018] The scaling module is used to connect the nodes sequentially to form a motion path polygon based on the spatial pose coordinates of each motion handover node in the basic motion feature set, calculate the ratio of the minimum feature radius to the area of ​​the motion path polygon, and scale the motion handover time impact rate sequence according to the ratio to obtain the corrected time impact rate sequence.

[0019] The correction module is used to perform time-series discrete accumulation on the corrected time impact rate sequence to obtain the overall fatigue cumulative effect value; based on the overall fatigue cumulative effect value, the working condition correction is performed on the basic time measurement values ​​in the basic working time parameter matrix to obtain the target working time data.

[0020] The above-described solution of the present invention has at least the following beneficial effects:

[0021] This system enables automated quantitative analysis of the entire process from discrete action codes to work time data. By matching the positional intervals of pure digital codes and retrieving standard parameters, it ensures the uniformity of basic action identification and the accuracy of data traceability. Through the action handover time impact rate and a dynamic correction mechanism based on the polygonal geometric features of the action path, it captures the physiological wear and spatial displacement resistance during action switching. Combined with the fatigue accumulation effect, it performs work condition adaptability calibration on basic work time, improving the on-site adaptability and dynamic response capability of work time setting. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating the standard working time dynamic parsing method based on a pure digital action code library provided in an embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram of a standard working time dynamic parsing system based on a pure digital action code library provided by an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram of the modified time impulse rate sequence provided in an embodiment of the present invention. Detailed Implementation

[0025] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0026] like Figure 1 As shown, embodiments of the present invention propose a standard working time dynamic parsing method based on a pure digital action code library, the method comprising the following steps:

[0027] Receive the pure numeric action code to be parsed, and convert the pure numeric action code to be parsed into an initial action instruction sequence according to the character position order;

[0028] Based on the initial action instruction sequence, the system matches the pre-set pure digital action code library according to the positional interval to obtain a basic action feature set composed of the action category interval corresponding to the first field and the detailed item sequence number corresponding to the subsequent fields.

[0029] Based on the detailed item sequence number in the basic motion feature set, the corresponding motion name and basic time measurement value are retrieved from the preset standard motion parameter benchmark table and summarized to form a basic working time parameter matrix.

[0030] Based on the basic working time parameter matrix, the instantaneous changes of adjacent basic time measurements at the action handover node are compared by difference to obtain the action handover time impact rate sequence.

[0031] Based on the spatial pose coordinates of each action handover node in the basic action feature set, the nodes are connected sequentially to form an action path polygon. The ratio of the minimum feature radius to the area of ​​the action path polygon is calculated. Based on the ratio, the action handover time impact rate sequence is scaled to obtain the corrected time impact rate sequence.

[0032] The corrected time impact rate sequence is subjected to time-series discrete accumulation operation to obtain the overall fatigue cumulative effect value; based on the overall fatigue cumulative effect value, the basic time measurement values ​​in the basic working time parameter matrix are corrected for the working conditions to obtain the target working time data.

[0033] In this embodiment of the invention, the entire chain of automated quantitative analysis from discrete action codes to work time data is realized. By matching the positional intervals of pure digital codes and retrieving standard parameters, the uniformity of basic action identification and the accuracy of data traceability are ensured. By using the action handover time impact rate and a dynamic correction mechanism based on the polygonal geometric features of the action path, the physiological wear and spatial displacement resistance during the action switching process are captured. In addition, the basic work time is calibrated for work condition adaptability in combination with the fatigue accumulation effect, thereby improving the on-site adaptability and dynamic response capability of the work time setting.

[0034] In a preferred embodiment of the present invention, the initial action instruction sequence may include:

[0035] In this embodiment, the pure numeric action code to be parsed is a sequence of numeric characters input from the front-end acquisition module of the standard time management system. The encoding rules for the pure numeric action code are predefined as follows: the total length of the code is a preset fixed number of digits, and each digit or each segment of consecutive digits corresponds to action information of different granularities. The preset fixed number of digits is based on the total number of action types and the number of detailed action items in the entire garment manufacturing process, and at the same time, a hierarchical matching mechanism is used to match the first field to distinguish the action category and the subsequent fields to match the action item. The fixed total length of the pure numeric action code to be parsed is uniformly set to 8 pure numeric characters. This number of digits is set according to the entire process operation scenario of the garment industry, such as sewing, cutting, finishing, and packaging, and can fully cover the eight major operation categories of material picking, displacement, positioning, sewing, adjustment, material feeding, resting, and auxiliary operation. Specifically, the conversion process includes the following operations: receiving the pure numeric action code to be parsed through the data interface of the standard time management system. This pure numeric action code to be parsed can come from the process action code output by the process design module or from the operation action code entered in real time by the on-site acquisition terminal. The received pure numeric action code is temporarily stored in the system memory in the form of a string.

[0036] According to the preset character positional division rules, the core basis is to first distinguish the major action categories and then match the subdivided actions. Combined with the fixed 8-bit code length, the positional segmentation parsing rules are formulated. Specifically, starting from the leftmost (i.e., the highest bit) of the code string, the value of each bit is read in sequence. The value of each bit and its position number in the code string together form an action instruction unit. Each action instruction unit contains the following two elements: first, the position number, which indicates the position order of the instruction in the original code string; second, the numeric value at that position, which indicates the action encoding information corresponding to that position. When the pure numeric action code to be parsed is a string composed of multiple digits, it is scanned from the first bit to the last bit. Each scan generates an instruction entry containing a position number and a numeric value. All instruction entries corresponding to the position are arranged in ascending order of position, which constitutes the initial action instruction sequence. This initial action instruction sequence is stored in the system memory as an ordered list data structure. Each item in the list is an instruction pair (position, value). The order of the items in the list strictly maintains the left-right arrangement order of the characters in the original code string.

[0037] In a preferred embodiment of the present invention, the basic action feature set may include:

[0038] The pure digital motion code library is a pre-built motion code dictionary stored in the system database. In this code library, all motion codes are organized in a hierarchical structure. The top level is the motion category, and each motion category has several motion sub-items. The motion category refers to the macro classification of work actions, such as "material picking and placing", "machine operation", "sewing operation", "quality inspection action", etc. Each motion sub-item is the specific operation step under the corresponding motion category, such as "picking up the cut piece with the left hand", "placing the cut piece with the right hand", "carrying the fabric with both hands" under the "material picking and placing" category.

[0039] The initial action instruction sequence is split into a first field and subsequent field sequences based on field length. Field length refers to the rule for dividing the number of digits used to distinguish between action categories and action sub-items, as defined in the pure digital action code library encoding rules. The preset field length parameter is based on the aforementioned 8-digit fixed-length pure digital action code encoding specification, while also adapting to the classification quantity and sub-item encoding capacity requirements of the four core operation categories in garment discrete manufacturing. In this embodiment, the preset field length parameter is 1 digit, meaning that a single field occupies a fixed 1-digit numeric character, which is uniformly used as the hierarchical splitting benchmark. The highest bit (i.e., the leftmost bit) of the code is a 1-digit fixed-length first field, which is dedicated to identifying the action category; in addition to the first... The remaining 7 digits outside the first field are evenly divided into 1-digit standard field units to form 7 subfields in the subsequent field sequence. Each subfield independently carries single-dimensional action detail encoding information. The specific splitting operation is as follows: read the initial action instruction sequence, and according to the preset field length parameter, extract the first few instruction units of the corresponding length from the beginning of the sequence and collect these instruction units into the first field. Continue to segment the remaining instruction units in the sequence according to the same field length parameter. Each segmentation result constitutes a subfield in the subsequent field sequence. After the splitting is completed, the first field (containing one field unit) and the subsequent field sequence (containing seven field units) are obtained.

[0040] Based on the value of the first field, the interval assignment is determined in the pre-defined action category interval mapping table of the pure digital action code library to obtain the action category interval corresponding to the first field. The action category interval mapping table is a core data table in the pure digital action code library. This table predefines the range of digital codes corresponding to each action category. Specifically, the mapping table is organized in the form of a triple of interval lower limit, interval upper limit, and action category name. Each row defines one action category. The intervals do not overlap and cover all possible value spaces (0-9) of the first field to ensure that the first field of any valid pure digital action code can match a unique action category. The specific value ranges of the interval lower limit and interval upper limit corresponding to each action category are as follows: Material handling category: interval lower limit 1, interval upper limit 2, covering the first field values ​​1 and 2, corresponding to material turnover actions such as cutting piece picking, fabric handling, and material placement; Machine operation category: interval lower limit 3, interval upper limit 5, covering the first field values ​​3, 4, and 5, corresponding to sewing machine start / stop, presser foot adjustment, and cutting knife. Equipment operation actions such as operation and equipment fine-tuning; sewing operation actions with a lower limit of 6 and an upper limit of 8, covering the first and second field values ​​of 6, 7, and 8, corresponding to core sewing actions such as straight sewing, overlocking, binding, point fixing, and corner sewing; quality inspection actions with a lower limit of 9 and an upper limit of 9, covering the first and second field value of 9, corresponding to quality inspection actions such as cut piece inspection, finished product quality inspection, defect investigation, and workstation self-inspection; reserved auxiliary actions with a lower limit of 0 and an upper limit of 0, covering the first and second field value of 0, corresponding to workstation tidying, thread trimming, and table maintenance. Cleaning and other auxiliary rest actions are used to adapt to the auxiliary operation of the process. The interval classification determination operation is as follows: the system extracts the value of the first field and compares the value with the lower limit and upper limit of the interval of each row in the action category interval mapping table. When the value is greater than or equal to the lower limit of the interval of a certain row and less than or equal to the upper limit of the interval of that row, it is determined that the first field belongs to the action category corresponding to that row. After the determination is completed, the system records the name of the action category and the corresponding interval identifier as the highest level classification result of the action code to be parsed.

[0041] Based on the values ​​of each field in the subsequent field sequence, sequence number matching is performed within the sub-item coding subset under the action category range to obtain the sub-item sequence number corresponding to each subsequent field. In the pure numeric action code library, each action category has a sub-item coding subset, which is a mapping table with the sub-item sequence number as the primary key. Each record in the table contains the sub-item sequence number (numeric code), the sub-item name, and the additional attribute parameters of the sub-item. The encoding rules of the sub-item sequence number are aligned with the numeric format of each field in the subsequent field sequence. Therefore, the value of each field in the subsequent field sequence can be directly used as a search keyword. The matching operation is as follows: The system retrieves the value of each field sequentially according to the order of the fields in the subsequent field sequence. For each field value, the system performs a matching query in the defined sub-item code subset of the current action category. That is, it uses the field value as the search condition to perform an equality search in the sub-item sequence number column of the sub-item code subset. If a matching record is found, the sub-item sequence number, sub-item name, and other information in the record are extracted and associated with the position of the field in the subsequent field sequence. If no matching record is found, a null value flag is returned and an exception log is recorded, indicating that the field value does not have a corresponding sub-item definition under the current action category.

[0042] The action category ranges are structurally associated with all detailed item sequence numbers to form a basic action feature set with the action category as the main index and the detailed item sequence numbers as sub-indexes. This structural association means assembling the obtained action category range results and the obtained detailed item sequence number results into a unified data structure according to hierarchical relationships. Specifically, the basic action feature set is implemented using a nested data structure. The outer record contains an "Action Category" field (main index), whose value is the specific action category name and range identifier. The inner record contains a "Detailed Item List" field, where each item is a detailed item entry (sub-index). Each detailed item entry contains a detailed item sequence number, a detailed item name, and its position in the subsequent field sequence. The arrangement order of each detailed item entry in the basic action feature set is consistent with the original arrangement order of the corresponding fields in the subsequent field sequence. This basic action feature set completely describes the hierarchical feature structure of "Action Category + Dimensional Detailed Items" corresponding to the input pure numeric action code.

[0043] In a preferred embodiment of the present invention, the basic working time parameter matrix may include:

[0044] The Standard Motion Parameter Baseline Table is another core data table in the pure digital motion code library. This table pre-includes standardized parameter information corresponding to each motion sub-item. Specifically, the Standard Motion Parameter Baseline Table uses the sub-item sequence number as the primary key, with each row corresponding to a motion sub-item. Each column stores the following attributes of the motion sub-item: motion name (human-readable motion description text), base time measurement value (standard time consumption measured in time units), and the spatial pose coordinate dataset associated with the motion sub-item (used for path analysis). The base time measurement value refers to the baseline time consumption of the motion sub-item determined by taking a predetermined number of repeated measurements and statistically averaging the measurement results under standard operating conditions. Standard operating conditions include standard operator skill level, standard workstation layout, and standard material status.

[0045] The system iterates through each sub-item sequence number in the basic motion feature set, using each sub-item sequence number as the retrieval primary key. A matching query is then performed in the pre-defined standard motion parameter benchmark table to extract the standard motion name and standard basic time measurement value corresponding to each sub-item sequence number. The traversal operation proceeds sequentially from the sub-item list of the basic motion feature set. For each sub-item sequence number, the system initiates a matching query in the standard motion parameter benchmark table using that sub-item sequence number as the retrieval primary key. Since the standard motion parameter benchmark table has a database index built using the sub-item sequence number as the primary key, this query operation can be completed in constant time. When the query returns a matching record, the system extracts the value of the "Action Name" field from the record and records it as the standard action name of the current sub-item; it extracts the value of the "Base Time Measurement Value" field and records it as the standard base time measurement value of the current sub-item; at the same time, it extracts the value of the "Spatial Pose Coordinates" field and records it as the spatial pose coordinate data of the current sub-item, and appends it to the attribute set of the sub-item for use in spatial path analysis. If the query does not return any matching record, that is, the sub-item sequence number does not exist in the standard action parameter baseline table, the system records the abnormal information, skips the sub-item, and continues to process the next sub-item sequence number.

[0046] The extracted standard action names and standard basic time measurements are arranged according to their original position in the basic action feature set, and combined to obtain a basic work-hour parameter matrix including an action identifier field and a time quantity field. The basic work-hour parameter matrix is ​​a two-dimensional data structure, where each row corresponds to a specific item in the basic action feature set. The row order strictly maintains the original position of the specific item in the basic action feature set. Specifically, the structure of the basic work-hour parameter matrix is ​​defined as follows: the number of rows in the matrix is ​​the total number of specific items in the basic action feature set, denoted as N, where N is a positive integer not less than 2. The matrix contains two columns: the first column is the "Action Identifier" field, storing the standard action name (text type) of the corresponding specific item; the second column is the "Time Quantity" field, storing the... The matrix contains the standard baseline time measurement values ​​(numerical type, unit is time measurement unit) for each action item. The element in the first column of the i-th row represents the standard action name of the i-th action item, and the element in the second column of the i-th row represents the standard baseline time measurement value of the i-th action item. In addition, each row has an additional hidden attribute column that stores the spatial pose coordinate data corresponding to the action item. This hidden column does not directly participate in the time calculation, but it will be referenced in the spatial path analysis. After the basic time parameter matrix is ​​assembled, it will be used as the input data for time impact analysis. It should be noted that the arrangement order of each row in the basic time parameter matrix already implies the temporal sequence of action execution. The first row corresponds to the first action item executed in the process, the second row corresponds to the second action item executed, and so on up to the Nth row.

[0047] In a preferred embodiment of the present invention, the action handover time impact rate sequence may include:

[0048] In this embodiment, the action handover node refers to the switching time between two consecutive action items in the process. The end time of the previous action item is also the start time of the next action item. In an ideal standard working time relationship, there is no gap or overlap between adjacent actions, and the end time of the previous action and the start time of the next action are the same point on the time axis. However, in actual operation, due to the switching of the operator's limb movement state, such as switching from a large displacement action to a high-precision positioning action, additional time consumption will occur at this handover node. This time deviation caused by the sudden change in movement state is the action handover time impact.

[0049] To quantify the time impact effect, the differences between the basic time measurements of adjacent actions are analyzed. The specific implementation process includes: extracting each pair of adjacent basic time measurements from the basic time parameter matrix according to the temporal sequence of the action handover nodes; determining the time measurement value corresponding to the end time of the previous action and the time measurement value corresponding to the start time of the next action for each pair of adjacent basic time measurements; and proceeding the extraction operation pairwise along the row order of the basic time parameter matrix (from top to bottom). Specifically, for two adjacent rows in the basic time parameter matrix, the k-th row and the (k+1)-th row, where k takes any integer value from 1 to N-1, the row position between these two rows is the k-th action handover node. The system extracts the value of the "time quantity" field in the k-th row and uses it as the relevant time measurement value of the previous action (the k-th action); simultaneously, it extracts the value of the "time quantity" field in the (k+1)-th row and uses it as the relevant time measurement value of the next action (the (k+1)-th action).

[0050] Since the basic time parameter matrix stores the standard basic time measurement values ​​of each action item, that is, the total time of each action, rather than the start or end time of the action, the time measurement value of the end time of the previous action and the time measurement value of the start time of the next action are conceptual terms in the standard time model. Under the assumption of continuous action in the standard time relationship, the end of the previous action and the start of the next action coincide at the same point on the time axis. Therefore, these two time measurements actually point to the standard basic time measurement values ​​of adjacent actions, and they together describe the time consumption characteristics of the two actions before and after the handover node.

[0051] The system calculates the difference between the time measurement at the start of the next action and the time measurement at the end of the previous action to obtain the instantaneous time change at each action handover node. For the k-th action handover node, the system subtracts the standard base time measurement of the previous action (k-th action) from the standard base time measurement of the next action (k+1-th action), and records the result as the instantaneous time change at the k-th action handover node. The physical meaning of this difference calculation is that the greater the difference in time consumption between two adjacent actions, the more significant or complex the motion state switch the operator needs to complete at the handover node, switching from a longer action mode to a shorter action mode, or vice versa. The time impact effect generated by acceleration or deceleration, posture adjustment, and other operations during this switching process is also more significant. The sign of the difference indicates the switching direction, i.e., a negative value from long time consumption to short time consumption, and a positive value from short time consumption to long time consumption; the absolute value of the difference reflects the intensity of the switch.

[0052] The instantaneous time changes are arranged chronologically according to the action handover nodes, and dimensionless ratio processing is performed using the baseline time measurement value of the corresponding action in the basic time parameter matrix as a reference. This yields the action handover time impact rate sequence. The purpose of dimensionless processing is to eliminate the dimensional influence caused by the difference in baseline time consumption between different action details, so that the time impact effects at different handover nodes can be compared and subsequently accumulated on the same scale. The specific operation of dimensionless processing is as follows: for the instantaneous time change at the k-th action handover node, this instantaneous time change is denoted as... ;The standard base time measurement value of the next action (i.e., the (k+1)th action) involved in the handover node (denoted as ); Using this as a reference benchmark, ratio calculations are performed. Specifically, the ratio is calculated as follows: absolute value and Perform a division operation and record the resulting quotient as the time impact rate value of the k-th action handover node. This ratio is a dimensionless ratio value, representing the proportion of the time change at the action handover point to the standard time consumption of the subsequent action. After performing the above ratio operation on all N-1 action handover nodes, arrange the ratios in order from node number 1 to N-1 to form the action handover time impact rate sequence. The length of this sequence is N-1, which is consistent with the number of action handover nodes.

[0053] like Figure 3 As shown, in a preferred embodiment of the present invention, the modified time impulse rate sequence may include:

[0054] In the actual operation of discrete garment manufacturing, the movement trajectory of the operator's two hands or one hand in conjunction with the tool in the workstation space constitutes a three-dimensional spatial path. The positional distribution of each action handover node in space, especially the turning angle, smoothness and coverage of the path, has a significant impact on the time impact effect when switching actions. The more tortuous the path and the larger the turning angle, the more drastic the posture adjustment and speed direction change that the operator needs to make at the handover node, and the greater the time impact generated.

[0055] In this embodiment, spatial pose coordinates refer to a six-dimensional data set describing the position and orientation of the operator's hand or end effector in the three-dimensional space of the workstation. It includes three position components (X-axis position component, Y-axis position component and Z-axis position component) and three attitude components (pitch angle, yaw angle and roll angle). Each action handover node is bound to a set of spatial pose coordinates. This coordinate data is pre-collected by motion capture equipment or manual calibration when the action details are entered into the standard action parameter benchmark table.

[0056] The first sub-stage involves constructing the spatial path polygon. The spatial normal vector parameters of the standard operation reference plane are pre-calibrated according to the workstation layout during system initialization. For a horizontal workbench, the normal vector direction is vertically upward; for a vertical work panel, the normal vector direction is horizontally pointing towards the operator. The orthogonal mapping relationship refers to the vertical projection relationship from any point in three-dimensional space to the reference plane. The projection direction is parallel to the normal vector direction of the reference plane, and the projection point is the closest point of that three-dimensional point on the reference plane. This mapping relationship is determined by the plane equation of the standard operation reference plane. Specifically, the reference plane is completely determined by the coordinates of a known point on it and the normal vector direction. The coordinates of the known point are selected during the system calibration stage from the corresponding point on the reference plane of the preset geometric center of the workstation operation area.

[0057] The preset geometric center is based on the standard sewing workstation table size in the garment industry (the effective working area of ​​a standard table is 1200mm long and 600mm wide). Using the geometric center of the effective working area as the reference point can maximize the coverage of all hand movements during fabric picking, sewing, and finishing, avoiding overall deviation of the projected coordinates caused by reference offset. This ensures that the geometric calculation reference of all work path polygons is unified and reusable, conforming to the characteristic of centered distribution of manual flexible work trajectories. The preset geometric center is the rectangular geometric center of the effective working table of the standard garment workstation, with three-dimensional calibration coordinates of (600mm, 300mm, 800mm). X and Y correspond to the horizontal two-dimensional coordinates of the table, and Z corresponds to the height of the standard working table. This point is the only fixed calibration reference point of the system and is uniformly reused for all processes and styles. For the three-dimensional position coordinates of each action handover node, the coordinates are composed of position components in the X-axis direction, Y-axis direction, and Z-axis direction. The system calculates the projection coordinates of the three-dimensional point on the reference plane along the normal vector direction according to the determined orthogonal mapping relationship. The projection coordinates are represented in the form of two-dimensional planar coordinates, where the first coordinate value represents the coordinate component along the first coordinate axis in the two-dimensional local coordinate system established on the reference plane, and the second coordinate value represents the coordinate component along the second coordinate axis in the two-dimensional local coordinate system. The preset coordinate origin is based on the reference origin for calculating the area of ​​a unified planar polygon, avoiding the deviation of positive and negative polygon areas caused by random origins, ensuring that the geometric feature parameters of different processes and different action paths are comparable laterally, and adapting to the standardized time analysis requirements of multiple styles and processes of clothing. The preset coordinate origin completely coincides with the preset geometric center of the above-mentioned workstation, that is, the coordinates of the origin of the two-dimensional local coordinate system are (600mm, 300mm), which is consistent with the horizontal projection coordinates of the three-dimensional geometric center and remains fixed throughout the process.

[0058] The origin of the two-dimensional local coordinate system is taken as the preset geometric center of the workstation operation area. The direction of the first coordinate axis is taken as the direction of the long side of the worktable, and the direction of the second coordinate axis is taken as the direction of the short side of the worktable. After performing vertical projection transformation on all N action intersection nodes one by one, the projection coordinates are arranged according to the original position order of the nodes in the basic action feature set to obtain the planar projection coordinate sequence.

[0059] Based on the position of the action handover nodes in the basic action feature set, adjacent planar projection coordinates are connected by vector lines to form a closed planar action path polygon. The connection operation starts from the first node, and adjacent planar projection coordinate points are connected by straight line segments in sequence. The first node is connected to the second node, the second node is connected to the third node, and so on, until the Nth node is connected back to the first node, so that the path is connected end to end. The resulting closed polygon is the planar action path polygon. This polygon is enclosed by N boundary line segments. The two endpoints of the i-th boundary line segment are the i-th node and the (i+1)-th node, respectively. When i is N, the (N+1)-th node returns to the first node, forming a closed loop. The geometry of this polygon intuitively reflects the projection coverage, number of turns, and regularity of the path of the operator's hand movement trajectory on the operating table during the entire process.

[0060] Based on all planar projection coordinates of the polygonal motion path, calculate the geometric centroid coordinates of the polygon and construct perpendicular lines from the geometric centroid coordinates to each boundary line segment of the polygonal motion path. The geometric centroid coordinates are calculated by taking the arithmetic mean of the first coordinate axis components of all N nodes and recording the average value as the first coordinate axis component of the centroid. The second coordinate axis components of all N nodes are also taken as the arithmetic mean of the average value, and the coordinates formed by the first and second coordinate axis components of the centroid are the geometric centroid coordinates. For each boundary line segment of the polygon, i.e., the line segment determined by the two endpoints of the i-th node and the (i+1)-th node, the system calculates the perpendicular foot coordinates from the centroid coordinates to the line containing the boundary line segment and constructs a perpendicular line segment with the centroid and the perpendicular foot as endpoints.

[0061] The length of the perpendicular segment corresponding to each boundary line segment is calculated, and the minimum length value is determined as the minimum feature radius of the planar motion path polygon. The length of the perpendicular segment is the perpendicular distance from the centroid to the corresponding boundary line segment, which in geometry is the distance from a point to a line. The system calculates the length of the perpendicular segment for each of the N boundary line segments of the polygon. Specifically, for each boundary line segment, the coordinates of the centroid and the coordinates of the two endpoints of the boundary line segment are used as input. The solution is performed according to the standard geometric relationship of the distance from a point to a line. That is, the area of ​​the triangle is calculated twice using these three points as vertices. Then, this value is divided by the length of the line segment between the two endpoints of the boundary line segment. The quotient is the length of the perpendicular segment. The minimum value among all N perpendicular segment lengths is determined as the minimum feature radius of the planar motion path polygon.

[0062] The physical meaning of the minimum feature radius is the distance from the nearest boundary to the center when viewed from the geometric center of the polygon. This value reflects whether there is a "bottleneck" in the polygon. The smaller the minimum feature radius, the more significant the "narrowing" part in the path. When the operator switches actions near the narrowing area, the less space there is for flexible hand movement.

[0063] The planar projection coordinates of adjacent polygons in the planar motion path are extracted according to position order. The area of ​​the triangle formed by the adjacent planar projection coordinates and the preset coordinate origin is calculated, and the areas of all triangles are accumulated to obtain the geometric area of ​​the planar motion path polygon. In this embodiment, the preset coordinate origin is the origin of the two-dimensional local coordinate system, that is, the preset geometric center of the workstation operation area. The area calculation is performed using the following segmentation steps: For the i-th boundary line segment, the coordinates of its endpoints are the planar projection coordinates of the i-th node and the planar projection coordinates of the (i+1)-th node, respectively. The product of the first coordinate axis component of the i-th node and the second coordinate axis component of the (i+1)-th node is denoted as the first intersection component. The product of the first coordinate axis component of the (i+1)th node and the second coordinate axis component of the ith node is denoted as the second cross component; the difference between the first and second cross components is denoted as the single-sided coordinate cross difference; the product of the single-sided coordinate cross difference and half of it is denoted as the sub-area corresponding to the ith boundary segment; for all N boundary segments, i is taken sequentially from 1 to N, and steps S55a to S55d are executed one by one to obtain N sub-area values. It should be noted that when i is N, the (N+1)th node is the first node, and the corresponding first and second coordinate axis components are respectively taken from the coordinate components of the first node, thereby completing the closure of the polygon boundary.

[0064] The summation of all N sub-area values ​​is performed, and the absolute value of the summation result is recorded as the geometric area of ​​the polygon of the planar motion path. The geometric principle of this area calculation method is to decompose the polygon into N triangles with the preset coordinate origin as the common vertex. Each triangle is composed of the coordinate origin and two adjacent vertices of the polygon. The sub-area of ​​each triangle is calculated by the coordinate cross difference method described above. The total geometric area of ​​the polygon is obtained by taking the absolute value of the sum of the sub-areas of all triangles.

[0065] The ratio of the minimum feature radius to the geometric area is calculated to obtain the ratio of the minimum feature radius to the area of ​​the motion path polygon. This ratio describes the compactness of the motion path. Specifically, the determined minimum feature radius is used as the dividend, and the obtained geometric area is used as the divisor. The resulting quotient is the ratio of the minimum feature radius to the area of ​​the motion path polygon. The dimension of this ratio is the reciprocal of length. Its physical meaning is that, with the same geometric area, the larger the ratio, the larger the minimum feature radius, the more full and smooth the path shape tends to be. The distance from the geometric center of the polygon to any boundary is relatively balanced, and there are no excessively narrow areas. The smaller the ratio, the smaller the minimum feature radius, the more elongated the path shape tends to be or there are obvious narrowing parts. The operator's activity space in certain areas of the path is restricted, and the changes in direction and speed adjustments during motion transitions are more drastic.

[0066] The second sub-stage scaling correction of the time impact rate involves extracting the ratio of the minimum feature radius to the area of ​​the motion path polygon, calculating the quotient of this ratio with the preset standard path benchmark ratio, and obtaining the path smoothness characterization parameter. The standard path benchmark ratio is a reference value pre-calculated and stored by the system during the initialization phase based on the ideal work path. The ideal work path refers to the geometric state of the path when all action junction nodes are evenly distributed on a circular path. In this state, the smoothness of the path reaches its theoretical limit. This benchmark ratio reflects the radius-area ratio characteristic that a final smooth path should have. The preset standard path benchmark ratio is based on the final circular work trajectory of the garment workstation as the theoretical benchmark. The circle has the characteristics of final geometric symmetry, no turning points, and no narrowing bottlenecks, making it an ideal smooth path for manual flexible operations. It can be used as a comparison benchmark for all actual work paths to achieve normalized and quantitative comparison of path smoothness. Based on the effective work area of ​​the standard workstation, the final circular work path is fitted, and the standard path benchmark ratio is fixed after geometric calculation. The system stores the data permanently and calls it globally. The specific calculation operation is to use the ratio of the minimum feature radius to the area of ​​the obtained action path polygon as the dividend and the preset standard path benchmark ratio as the divisor, and perform a division operation. The resulting quotient is recorded as the path smoothness characterization parameter. When the parameter is greater than 1, it means that the actual path is more compact and full than the ideal path. When the parameter is close to 1, it means that the smoothness of the actual path is close to that of the ideal path. When the parameter is less than 1, it means that the actual path has obvious narrowing or twisting, and the smoothness is lower than the ideal level.

[0067] Based on the path smoothness characterization parameters, a matching query is performed in the preset loss compensation mapping relationship to obtain the time impact scaling factor. The loss compensation mapping relationship is a pre-built mapping data structure in the system. This mapping relationship describes the correspondence between path smoothness and the scaling compensation required for the time impact effect it produces. Specifically, the absolute value of the difference between the path smoothness characterization parameters and the preset standard smoothness benchmark value is calculated to obtain the smoothness deviation feature value. The standard smoothness benchmark value is a preset reference constant with a value of 1, corresponding to the theoretical smoothness level of the final smoothed path. This benchmark value is used to measure the gap between the actual path smoothness and the ideal level. The calculation method is to subtract the standard smoothness benchmark value (i.e., the value 1) from the path smoothness characterization parameters and take the absolute value of the subtraction result. The resulting absolute value is recorded as the smoothness deviation feature value. The larger this feature value is, the greater the deviation between the geometric characteristics of the actual operation path and the ideal smooth path, indicating that the operator needs to undergo more directional changes and posture adjustments when performing the operation.

[0068] Based on the smoothness deviation feature value, a matching process is performed within multiple deviation intervals included in the preset loss compensation mapping relationship to obtain the target deviation interval identifier corresponding to the smoothness deviation feature value. The loss compensation mapping relationship consists of multiple hierarchical deviation intervals, which are divided into several intervals in ascending order of deviation degree. Each deviation interval has a lower boundary value and an upper boundary value, and corresponds to a unique interval identifier. In this embodiment, combined with the measured data of the tortuosity of the action path in the garment process, the smoothness deviation feature value domain [0, 1.2] is divided into four equal layers. The specific lower boundary value and upper boundary value range of each deviation interval are as follows: the lower boundary of the slight deviation interval is 0.0, and the upper boundary is 0.2; the lower boundary of the mild deviation interval is 0.2. The lower boundary of the moderate deviation interval is 0.2, and the upper boundary is 0.5; the lower boundary of the moderate deviation interval is 0.5, and the upper boundary is 0.8; the lower boundary of the severe deviation interval is 0.8, and the upper boundary is 1.2. Each interval is continuous and connected without overlap or gaps, completely covering all effective smoothness deviation feature values, and corresponding to a unique interval identifier. The assignment matching operation is as follows: the system compares the smoothness deviation feature value with the lower boundary value and the upper boundary value of each deviation interval one by one. When the smoothness deviation feature value is greater than or equal to the lower boundary value of a certain interval and less than the upper boundary value of that interval, the interval identifier of that interval is determined as the target deviation interval identifier. Each deviation interval is continuous and without gaps, ensuring that any possible smoothness deviation feature value can be uniquely assigned to a deviation interval.

[0069] Based on the target deviation interval identifier, the corresponding baseline scaling weight is retrieved in the loss compensation mapping relationship, and the smoothness deviation feature value is used to perform gradient correction calculation on the baseline scaling weight to obtain the time impact scaling coefficient. In the loss compensation mapping relationship, each deviation interval is associated with a preset baseline scaling weight value. This baseline scaling weight is an empirically determined coefficient, representing the basic scaling factor to be applied to the time impact rate sequence at the baseline level of the deviation. The preset baseline scaling weight value is based on fitting of massive process time measurement data from the garment workshop, matching the action switching time impact loss amplitude corresponding to different path tortuosity, with a slight deviation interval [0, 0.2]. The ranges are: slightly off-range [0.2, 0.5), moderately off-range [0.5, 0.8), and severely off-range [0.8, 1.2]. The baseline scaling weight is 1.30~1.50. The preset linear interpolation step size within the range is based on the goal of achieving continuous gradient correction of the scaling weight within the range, avoiding abrupt changes in coefficients and correction values ​​at range nodes, ensuring a smooth transition of scaling coefficients affected by time impacts, improving the continuity and accuracy of time correction, and adapting to the fine-grained compensation requirements for subtle path geometric differences. The preset linear interpolation step size within the range is fixed at 0.01, meaning that for every 0.01 change in the normalized position ratio, the corresponding baseline scaling weight is simultaneously fine-tuned by 0.01 times the gradient increment.

[0070] The specific method for gradient correction calculation is as follows: within the target deviation interval, starting from the baseline scaling weight, calculate the relative position of the smoothness deviation feature value within the target deviation interval. First, subtract the lower boundary value of the target deviation interval from the smoothness deviation feature value to obtain a difference. Then, divide this difference by the interval width of the target deviation interval (i.e., the difference between the upper boundary value and the lower boundary value) to obtain a normalized position ratio. According to the relationship between the preset linear interpolation step size within the interval and this normalized position ratio, incrementally correct the baseline scaling weight. The corrected weight value is determined as the time impact scaling coefficient. This coefficient is a dimensionless value, and its value is usually within an interval greater than or equal to 1. The larger the coefficient, the stronger the scaling compensation needs to be applied to the time impact rate sequence, that is, the further the path deviates from the ideal smooth state, and the more the corresponding action switching time impact effect needs to be amplified and reflected.

[0071] The system iterates through the action handover time impact rate sequence, multiplies each action handover time impact rate with a time impact scaling factor, and obtains an initial scaled time impact rate sequence. For each time impact rate value in the action handover time impact rate sequence, it is numbered k, where k ranges from 1 to N-1. The system multiplies the k-th time impact rate value with the time impact scaling factor, and records the product as the k-th initial scaled time impact rate. All initial scaled time impact rates are arranged according to their position in the original sequence to obtain the initial scaled time impact rate sequence. The physical meaning of this multiplication operation is that the time impact scaling factor acts as a global geometric-dynamic coupling amplification, mapping the spatial geometric quality of the path (described by the smoothness characterization parameter) to the overall scaling amount of the time impact effect. The worse the smoothness of the path, the larger the scaling factor, and the greater the amplification of the initial scaled time impact rate relative to the original time impact rate.

[0072] To avoid excessive amplification of the time impact rate due to extreme deviations in path geometry, the system performs upper limit threshold comparison and clipping operations on each initial scaled time impact rate. It extracts each initial scaled time impact rate from the initial scaled time impact rate sequence and compares each initial scaled time impact rate with a preset upper limit threshold for time impact. This upper limit threshold is an upper limit constant set during the system configuration phase based on the maximum statistical value of the time impact rate observed in the actual production environment. It constrains the reasonable range of time impact rate values. The preset upper limit threshold is based on statistical data of on-site measured working hours from high-frequency action switching and extreme posture adjustments throughout the garment manufacturing process. It filters out effective time impact rate extremes within a 99.7% confidence interval, eliminating invalid excessively large values ​​caused by extreme abnormal trajectories, preventing distortion of fatigue cumulative effect values ​​and working hour correction distortion, and ensuring the algorithm's robustness matches actual production conditions. The preset upper limit threshold range is [1.8, 2.0]. In this embodiment, a fixed threshold of 2.0 is used as the sole upper limit, and all scaled impact rates do not exceed this value.

[0073] For each initial scaling time impact rate, if its value is less than or equal to the upper limit threshold of time impact, no pruning operation is performed, and the initial scaling time impact rate is directly used as the corresponding corrected time impact rate. For each initial scaling time impact rate, if its value is greater than the upper limit threshold of time impact, the value of the upper limit threshold of time impact is used as the corresponding corrected time impact rate, that is, the threshold is used to replace the out-of-bounds value. After performing the above judgment and pruning operations on all N-1 initial scaling time impact rates in sequence, all corrected time impact rates are arranged according to the original position order of the initial scaling time impact rate sequence, thus obtaining the corrected time impact rate sequence. Each item in this sequence is constrained to a reasonable range not exceeding the upper limit threshold of time impact, and is a sequence of time impact rate values ​​corrected by spatial path geometric features.

[0074] In a preferred embodiment of the present invention, the target working time data may include:

[0075] The modified time impact rate sequence is discretized and accumulated item by item according to the action execution sequence to obtain the overall fatigue cumulative effect value. The modified time impact rate sequence contains N-1 values, arranged sequentially from 1 to N-1, and the arrangement order strictly follows the temporal order of the action handover nodes. The specific operation of discretization and item-by-item accumulation is as follows: initialize an accumulator variable and set its value to zero; then, in ascending order of the number, take each value in the modified time impact rate sequence, add the current value of the accumulator variable to the selected modified time impact rate value each time, and use the result of the addition operation as the new value of the accumulator variable; after all N-1 items are added, the final value of the accumulator variable is the overall fatigue cumulative effect value.

[0076] The physical meaning of the overall fatigue cumulative effect value is the total cumulative time impact experienced by the operator in a complete process cycle, from the first action handover node to the last action handover node. Since each corrected time impact rate is a dimensionless ratio value, representing the proportion of extra time consumption at that handover node relative to the standard baseline time, their sum characterizes the relative amount of total time deviation caused by the dynamic effect of action connection in the entire process. The larger the value, the more significant the dynamic impact effect of action connection in the process, and the higher the degree of fatigue accumulation of the operator.

[0077] Using the overall fatigue cumulative effect value as the working condition correction coefficient, the basic time measurement value corresponding to each action in the basic working time parameter matrix is ​​corrected to obtain the corrected time measurement value of each action. Working condition correction refers to adjusting the basic time measurement value under standard working conditions to reflect the actual working conditions, that is, the corrected time value including the dynamic effect of action connection and fatigue cumulative effect. The specific correction operation is as follows: for the basic time measurement value of the i-th action in the basic working time parameter matrix, where i ranges from 1 to N; perform the correction calculation according to the segmentation step, add the value 1 with the overall fatigue cumulative effect value, and record the sum as the comprehensive correction coefficient; multiply the basic time measurement value of the i-th action with the comprehensive correction coefficient, and record the product as the corrected time measurement value of the i-th action; take i from 1 to N, and perform the above calculation one by one to obtain the corrected time measurement value of all N actions.

[0078] The meaning of this correction method is that the base time measurement reflects the standard time consumption under ideal continuous action conditions; the comprehensive correction coefficient reflects the proportion of additional time consumption caused by the cumulative dynamic impact effect of all handover nodes in the entire action sequence, where a value of 1 corresponds to the ideal situation with no additional consumption, and the overall fatigue cumulative effect value corresponds to the accumulated additional consumption proportion; when the overall fatigue cumulative effect value is zero (i.e., ideal transition is achieved at all handover nodes with no time impact), the comprehensive correction coefficient is equal to a value of 1, and the corrected time measurement value is equal to the base time measurement value, that is, the working condition of the current process is consistent with the standard working condition, and no correction is required; when the overall fatigue cumulative effect value is greater than zero, the comprehensive correction coefficient is greater than a value of 1, and the corrected time measurement value is greater than the base time measurement value, that is, positive time compensation is performed based on the fatigue cumulative effect.

[0079] The system summarizes all corrected time measurements according to the original action sequence to obtain the target time data. The system extracts the corrected time measurement value and the corresponding action name of each action according to the row order in the basic time parameter matrix, i.e. the original action execution sequence, and summarizes them to form the target time data table. Each row of the target time data table contains three fields: action sequence (the execution sequence number of the action in the process), action name (the standard action name of the action), and corrected time measurement value (the time measurement value of the action after working condition correction, in time measurement units). The target time data is the collection of all corrected time measurement values ​​in this data table, which can be directly used for downstream applications such as standard time setting of the process, production line balancing calculation, and production scheduling.

[0080] like Figure 2 As shown, embodiments of the present invention also provide a standard working time dynamic parsing system based on a pure digital action code library, including:

[0081] The parsing module is used to receive the pure numeric action code to be parsed and convert the pure numeric action code into an initial action instruction sequence according to the character position order;

[0082] The matching module is used to match the initial action instruction sequence in a preset pure digital action code library according to the positional interval, and obtain a basic action feature set composed of the action category interval corresponding to the first field and the detailed item sequence number corresponding to the subsequent fields.

[0083] The retrieval module is used to retrieve the corresponding action name and basic time measurement value from the preset standard action parameter benchmark table based on the detailed item sequence number in the basic action feature set, and summarize them to form a basic working time parameter matrix.

[0084] The calculation module is used to compare the instantaneous changes of adjacent basic time measurements at the action handover node based on the basic time parameter matrix to obtain the action handover time impact rate sequence.

[0085] The scaling module is used to connect the nodes sequentially to form a motion path polygon based on the spatial pose coordinates of each motion handover node in the basic motion feature set, calculate the ratio of the minimum feature radius to the area of ​​the motion path polygon, and scale the motion handover time impact rate sequence according to the ratio to obtain the corrected time impact rate sequence.

[0086] The correction module is used to perform time-series discrete accumulation on the corrected time impact rate sequence to obtain the overall fatigue cumulative effect value; based on the overall fatigue cumulative effect value, the working condition correction is performed on the basic time measurement values ​​in the basic working time parameter matrix to obtain the target working time data.

[0087] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0088] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A standard time dynamic parsing method based on a pure digital motion code library, characterized in that, The method includes: Step 1: Receive the pure numeric action code to be parsed, and convert the pure numeric action code to be parsed into an initial action instruction sequence according to the character position order; Step 2: Based on the initial action instruction sequence, match the preset pure digital action code library according to the position interval to obtain a basic action feature set composed of the action category interval corresponding to the first field and the detailed item sequence number corresponding to the subsequent fields. Step 3: Based on the detailed item sequence number in the basic motion feature set, retrieve the corresponding motion name and basic time measurement value from the preset standard motion parameter benchmark table, and summarize them to form a basic working time parameter matrix; Step 4: Based on the basic working time parameter matrix, compare the instantaneous changes of adjacent basic time measurements at the action handover node to obtain the action handover time impact rate sequence. Step 5: Based on the spatial pose coordinates of each action transition node in the basic action feature set, connect each node sequentially to form an action path polygon, and calculate the ratio of the minimum feature radius to the area of ​​the action path polygon; based on the ratio, scale the action transition time impact rate sequence to obtain the corrected time impact rate sequence. Step 6: Perform time-series discrete accumulation operation on the corrected time impact rate sequence to obtain the overall fatigue cumulative effect value; based on the overall fatigue cumulative effect value, perform working condition correction on the basic time measurement values ​​in the basic working time parameter matrix to obtain the target working time data.

2. The standard working time dynamic parsing method based on a pure digital action code library according to claim 1, characterized in that, Based on the initial action instruction sequence, matching is performed in a pre-defined pure digital action code library according to positional intervals to obtain a basic action feature set composed of the action category interval corresponding to the first field and the detailed item sequence number corresponding to the subsequent fields, including: The initial action instruction sequence is split into a first field and a subsequent field sequence according to the field length; Based on the value of the first field, the interval assignment is determined in the preset pure numeric action code library action category interval mapping table to obtain the action category interval corresponding to the first field. Based on the values ​​of each field in the subsequent field sequence, the sequence number is matched in the sub-item coding subset under the action category range to obtain the sub-item sequence number corresponding to each subsequent field; The action category range is structured and associated with all the detailed item sequence numbers to form a basic action feature set with the action category as the main index and the detailed item sequence number as the sub-index.

3. The standard working time dynamic parsing method based on a pure digital action code library according to claim 2, characterized in that, Based on the detailed item sequence number in the basic motion feature set, the corresponding motion name and basic time measurement value are retrieved from the preset standard motion parameter benchmark table, and summarized to form a basic time parameter matrix, including: Traverse each sub-item sequence number in the basic motion feature set, use each sub-item sequence number as the retrieval primary key, perform a matching query in the preset standard motion parameter benchmark table, and extract the standard motion name and standard basic time measurement value corresponding to each sub-item sequence number. The extracted standard action names and standard basic time measurement values ​​are arranged according to their original position in the basic action feature set according to the detailed item sequence number, and combined to obtain a basic working time parameter matrix including action identification field and time quantity field.

4. The standard working time dynamic parsing method based on a pure digital action code library according to claim 3, characterized in that, Based on the basic time parameter matrix, the instantaneous changes in adjacent basic time measurements at the action handover node are compared to obtain the action handover time impact rate sequence, including: According to the time sequence of the action handover nodes, each pair of adjacent basic time measurement values ​​in the basic time parameter matrix is ​​extracted in turn, and the time measurement value at the end of the previous action and the time measurement value at the start of the next action corresponding to each pair of adjacent basic time measurement values ​​are determined. The difference between the time measurement value at the start time of the next action and the time measurement value at the end time of the previous action is calculated to obtain the instantaneous time change at each action handover node. Arrange the instantaneous time changes according to the time sequence of the action handover nodes, and perform dimensionless ratio processing with the base time measurement value of the corresponding action in the base time parameter matrix as a reference to obtain the action handover time impact rate sequence.

5. The standard working time dynamic parsing method based on a pure digital motion code library according to claim 4, characterized in that, Based on the spatial pose coordinates of each action intersection node in the basic action feature set, the nodes are sequentially connected to form an action path polygon. The ratio of the minimum feature radius to the area of ​​the action path polygon is calculated, including: Extract the three-dimensional position components from the spatial pose coordinates of each action handover node in the basic action feature set; project the three-dimensional position components vertically onto the preset standard operation reference plane to obtain the planar projection coordinates of each action handover node in the standard operation reference plane; According to the position of the action handover node in the basic action feature set, the adjacent planar projection coordinates are connected by vectors to form a closed planar action path polygon. Based on all planar projection coordinates of the planar motion path polygon, calculate the geometric centroid coordinates of the polygon, and construct perpendicular lines from the geometric centroid coordinates to each boundary line segment of the planar motion path polygon. Calculate the length of the perpendicular line segment corresponding to each boundary line segment, and determine the minimum length value as the minimum feature radius of the planar motion path polygon; Extract the adjacent planar projection coordinates in the boundary of the planar motion path polygon according to the position order, calculate the area of ​​the triangle formed by the adjacent planar projection coordinates and the preset coordinate origin, and sum up the areas of all triangles to obtain the geometric area of ​​the planar motion path polygon. Calculate the quotient of the minimum feature radius and the geometric area to obtain the ratio of the minimum feature radius to the area of ​​the motion path polygon.

6. The standard working time dynamic parsing method based on a pure digital action code library according to claim 5, characterized in that, Extract the three-dimensional position components from the spatial pose coordinates of each action junction node in the basic action feature set; The three-dimensional position components are vertically projected onto a preset standard operating reference plane to obtain the planar projection coordinates of each action handover node within the standard operating reference plane, including: Based on the identifiers of each action handover node arranged in position in the basic action feature set, the spatial pose coordinate dataset bound to each node is retrieved by index, and the corresponding three-dimensional position components are separated from the spatial pose coordinate dataset. Based on each three-dimensional position component and combined with the spatial normal vector parameters of the preset standard operation reference plane, the orthogonal mapping relationship of each three-dimensional position component on the standard operation reference plane is determined. Based on the orthogonal mapping relationship, a vertical projection transformation is performed on each three-dimensional position component to obtain the planar projection coordinate sequence of each action handover node in the standard operation reference plane.

7. The standard working time dynamic parsing method based on a pure digital action code library according to claim 6, characterized in that, Based on the ratio, the action handover time impulse rate sequence is scaled to obtain a corrected time impulse rate sequence, including: Extract the ratio of the minimum feature radius to the area of ​​the motion path polygon, calculate the quotient of the ratio with the preset standard path benchmark ratio, and obtain the path smoothness characterization parameter; Based on the path smoothness characterization parameters, a matching query is performed in the preset loss compensation mapping relationship to obtain the time impact scaling factor. Iterate through each action handover time impact rate in the action handover time impact rate sequence, and multiply each action handover time impact rate with the time impact scaling factor to obtain the initial scaled time impact rate sequence. Extract each initial scaling time shock rate from the initial scaling time shock rate sequence, and compare each initial scaling time shock rate with a preset time shock upper limit threshold; If the initial scaling time shock rate is less than or equal to the upper limit threshold of the time shock, then the corresponding initial scaling time shock rate is used as the corrected time shock rate. If the initial scaling time shock rate is greater than the time shock upper limit threshold, then the time shock upper limit threshold is used as the corresponding corrected time shock rate. Arrange all the corrected time shock rates according to the position order of the initial scaled time shock rate sequence to obtain the corrected time shock rate sequence.

8. The standard working time dynamic parsing method based on a pure digital action code library according to claim 7, characterized in that, Based on the path smoothness characterization parameters, a matching query is performed within the preset loss compensation mapping relationship to obtain the time impact scaling factor, including: The absolute value of the difference between the path smoothness characterization parameter and the preset standard smoothness benchmark value is calculated to obtain the smoothness deviation characteristic value; Based on the smoothness deviation feature value, the assignment matching is performed in multiple deviation intervals included in the preset loss compensation mapping relationship to obtain the target deviation interval identifier corresponding to the smoothness deviation feature value. Based on the target deviation interval identifier, the corresponding baseline scaling weight is retrieved in the loss compensation mapping relationship, and the gradient correction calculation of the baseline scaling weight is performed on the smoothness deviation feature value to obtain the time impact scaling coefficient.

9. The standard working time dynamic parsing method based on a pure digital action code library according to claim 8, characterized in that, The overall fatigue cumulative effect value is obtained by performing a time-series discrete accumulation operation on the modified time-impact rate sequence. Based on the overall fatigue cumulative effect value, the basic time measurements in the basic time parameter matrix are corrected for working conditions to obtain the target working time data, including: The modified time impact rate sequence is discretized according to the action execution time sequence and accumulated item by item to obtain the overall fatigue cumulative effect value; The overall fatigue cumulative effect value is used as the working condition correction coefficient. The working condition correction is applied to the basic time measurement value corresponding to each action in the basic working time parameter matrix to obtain the corrected time measurement value of each action. All corrected time measurements are summarized according to the original action sequence to obtain the target working time data.

10. A standard time dynamic parsing system based on a pure digital motion code library, the system implementing the method as described in any one of claims 1 to 9, characterized in that, include: The parsing module is used to receive the pure numeric action code to be parsed and convert the pure numeric action code into an initial action instruction sequence according to the character position order; The matching module is used to match the initial action instruction sequence in a preset pure digital action code library according to the positional interval, and obtain a basic action feature set composed of the action category interval corresponding to the first field and the detailed item sequence number corresponding to the subsequent fields. The retrieval module is used to retrieve the corresponding action name and basic time measurement value from the preset standard action parameter benchmark table based on the detailed item sequence number in the basic action feature set, and summarize them to form a basic working time parameter matrix. The calculation module is used to compare the instantaneous changes of adjacent basic time measurements at the action handover node based on the basic time parameter matrix to obtain the action handover time impact rate sequence. The scaling module is used to connect the nodes sequentially to form a motion path polygon based on the spatial pose coordinates of each motion handover node in the basic motion feature set, calculate the ratio of the minimum feature radius to the area of ​​the motion path polygon, and scale the motion handover time impact rate sequence according to the ratio to obtain the corrected time impact rate sequence. The correction module is used to perform time-series discrete accumulation on the corrected time impact rate sequence to obtain the overall fatigue cumulative effect value; based on the overall fatigue cumulative effect value, the working condition correction is performed on the basic time measurement values ​​in the basic working time parameter matrix to obtain the target working time data.

Images