Working time data calculation method, device, computer equipment and storage medium

Abstract

The present invention provides a method, apparatus, computer device, and storage medium for calculating working time data. The method comprises parsing a work video to obtain process times for multiple processes; obtaining the process times of n objects in the same process to form process clusters; obtaining the cluster centers of the process clusters; calculating the Euclidean distance between each process time and the cluster center to obtain distance clusters; and calculating the optimal cluster center of the process cluster based on the normal distribution of the distance clusters using a maximum expectation algorithm. By collecting work videos during the production process, parsing the process times, and using a clustering algorithm to calculate the optimal cluster center, standard working time is obtained. This makes the calculation of standard working time more convenient and more accurate.

Description

Technical Field

[0001] The present invention relates to the technical field of man-hour calculation, and in particular to a method, device, computer equipment and storage medium for calculating man-hour data. Background Art

[0002] Currently, manufacturing companies generally have a large amount of basic working hours data, which is showing an increasing trend year by year, causing management problems. As manufacturing companies gradually shift their cost accounting methods to the working hours method, the requirements for working hours accuracy and detailed working hours standards are further improved. However, the current method of determining standard working hours mainly relies on manual statistics, which is inefficient and inaccurate. Therefore, it is very important to provide a convenient and scientific method for calculating and determining standard working hours. Summary of the Invention

[0003] Based on this, it is necessary to provide a method, device, computer equipment and storage medium for calculating working time data to address the above technical problems.

[0004] A method for calculating working time data, comprising:

[0005] Collect operation videos of preset workstations on the production line;

[0006] Analyze the operation video to obtain the process time of multiple processes;

[0007] Obtaining the process time of n objects of the same process to form the process cluster;

[0008] Obtaining the cluster center of the process cluster;

[0009] Calculating the Euclidean distance between each process time and the cluster center to obtain a distance accumulation cluster;

[0010] According to the normal distribution of the distance clusters, the optimal cluster center of the process clusters is calculated based on the maximum expectation algorithm.

[0011] In one embodiment, the step of collecting operation videos of preset workstations on the production line includes:

[0012] Collecting the operation video of the preset workstation on the production line;

[0013] Performing validity processing on the operation video to obtain a valid operation video;

[0014] The step of analyzing the operation video to obtain the process time of multiple processes includes:

[0015] The effective operation video is analyzed to obtain the process time of multiple processes.

[0016] In one embodiment, the step of performing validity processing on the operation video to obtain a valid operation video includes:

[0017] Get the preset slack factor and preset average working hours;

[0018] The operation video is cut based on the preset relaxation coefficient and the preset average working hours to obtain the effective operation video.

[0019] In one embodiment, the step of analyzing the operation video to obtain the process time of multiple processes includes:

[0020] Segmenting the operation video based on the process to obtain segmented videos of multiple processes;

[0021] Analyze each of the segmented videos to obtain the process time of multiple processes.

[0022] In one embodiment, after the step of calculating the optimal cluster center of the process cluster based on the normal distribution of the distance cluster using the maximum expectation algorithm, the method further includes:

[0023] Obtaining the correction process time calculated based on the newly collected operation video;

[0024] Based on the process time and the corrected process time of n objects in the same process, a corrected optimal cluster center is calculated.

[0025] In one embodiment, the steps of obtaining the corrected process time calculated based on the newly collected operation video, and calculating the corrected optimal cluster center based on the process time and the corrected process time of n objects of the same process are repeated until the offset of the corrected optimal cluster center is less than the preset offset.

[0026] A working time data calculation device, comprising:

[0027] The acquisition module is used to collect operation videos of preset workstations on the production line;

[0028] A process time acquisition module is used to analyze the operation video and obtain the process time of multiple processes;

[0029] A process cluster forming module is used to obtain n process times of the same process to form the process cluster;

[0030] A cluster center acquisition module, used to obtain the cluster center of the process cluster;

[0031] A distance cluster acquisition module is used to calculate the Euclidean distance between each process time and the cluster center to obtain a distance cluster;

[0032] The optimal cluster center obtaining module is used to calculate the optimal cluster center of the process cluster based on the maximum expectation algorithm according to the normal distribution of the distance cluster.

[0033] In one embodiment, the acquisition module includes:

[0034] An operation video acquisition unit, used to acquire the operation video of a preset workstation on the production line;

[0035] A validity processing unit, configured to perform validity processing on the operation video to obtain a valid operation video;

[0036] The process time acquisition module is used to analyze the effective operation video to obtain the process time of multiple processes.

[0037] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and wherein the processor performs the following steps when executing the computer program:

[0038] Collect operation videos of preset workstations on the production line

[0039] Analyze the operation video to obtain the process time of multiple processes;

[0040] Obtaining the process time of n objects of the same process to form the process cluster;

[0041] Obtaining the cluster center of the process cluster;

[0042] Calculating the Euclidean distance between each process time and the cluster center to obtain a distance accumulation cluster;

[0043] According to the normal distribution of the distance clusters, the optimal cluster center of the process clusters is calculated based on the maximum expectation algorithm.

[0044] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps:

[0045] Collect operation videos of preset workstations on the production line

[0046] Analyze the operation video to obtain the process time of multiple processes;

[0047] Obtaining the process time of n objects of the same process to form the process cluster;

[0048] Obtaining the cluster center of the process cluster;

[0049] Calculating the Euclidean distance between each process time and the cluster center to obtain a distance accumulation cluster;

[0050] According to the normal distribution of the distance clusters, the optimal cluster center of the process clusters is calculated based on the maximum expectation algorithm.

[0051] The above-mentioned working hour data calculation method, device, computer equipment and storage medium obtain the process time by collecting the operation video during the production process, and use the clustering algorithm to calculate the optimal cluster center to obtain the standard working hours, thereby making the calculation of the standard working hours more convenient and the calculated standard working hours more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 1 is a flow chart of a method for calculating working time data in one embodiment;

[0053] Figure 2 is a structural block diagram of a working time data calculation device in one embodiment;

[0054] Figure 3 is a diagram of the internal structure of a computer device in one embodiment;

[0055] Figure 4 Schematic diagram of a flow chart of a method for calculating working time data in another embodiment. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0057] Example 1

[0058] In this embodiment, Figure 1 As shown, a method for calculating working time data is provided, which includes:

[0059] Step 110: Collect operation videos of preset workstations on the production line.

[0060] Specifically, a camera is used to capture video of an operator at a workstation on the production line, thereby capturing a video of the work performed at that workstation. In this embodiment, the camera can capture work videos of the same operator at the workstation during different time periods, or it can capture work videos of different operators at different workstations during different time periods. In addition, in this embodiment, the line name and job title corresponding to the workstation are obtained. After obtaining the line name and job title, the line name and job title are associated with the work video.

[0061] Step 120: parse the operation video to obtain the process time of multiple processes.

[0062] In this embodiment, the operation video is parsed to obtain the duration of each process in the operation video. The duration of each process is the process time of the corresponding process, and the process time is the time required for the operator to complete the process. In this embodiment, the process time of multiple objects of multiple processes is obtained, each object is a worker. The process time can be obtained by parsing the operation video by obtaining the length of the operation video, or by using a machine learning model to determine the operator's process start time and process end time based on the operator's repeated operations in the operation video. This embodiment does not describe this in detail.

[0063] Step 130: Obtain the process time of n objects in the same process to form the process cluster.

[0064] In this step, the process times of multiple objects in the same process are obtained from the multiple process times obtained by analysis, that is, the process times of different operators in the same process are obtained, and these process times are integrated into a process cluster. In this embodiment, the process is X, and the process cluster of n objects formed is X={ , , ,..., }.

[0065] Step 140: Obtain the cluster center of the process cluster.

[0066] In this embodiment, the process cluster is X={ , , ,..., } to determine the cluster center of the cluster.

[0067] Step 150: Calculate the Euclidean distance between each process time and the cluster center to obtain a distance accumulation cluster.

[0068] In this embodiment, it is assumed that C is the cluster center of the process cluster, and the cluster { , , ,..., }, the cluster obtained by the calculation is the distance cluster.

[0069] Step 160 : According to the normal distribution of the distance clusters, the optimal cluster center of the process clusters is calculated based on the maximum expectation algorithm.

[0070] In this step, the optimal cluster center obtained by using the maximum expectation algorithm based on the normal distribution of the distance clusters is the required standard working hours.

[0071] In the above embodiment, by collecting the operation video during the production process, analyzing the process time, and using the clustering algorithm to calculate the optimal cluster center, the standard working hours are obtained, which makes the calculation of the standard working hours more convenient and the calculated standard working hours more accurate.

[0072] It should be understood that in the above steps 130 to 160, the k-means clustering algorithm is used to calculate the distance clusters, and the maximum expectation algorithm is combined to calculate the optimal cluster center of the process clusters, thereby obtaining the standard working hours, thereby calculating the standard working hours of the process, and using this to construct a standard working hour determination model.

[0073] In one embodiment, the step of collecting the operation video of the preset workstation on the production line includes: collecting the operation video of the preset workstation on the production line; performing validity processing on the operation video to obtain a valid operation video; the step of parsing the operation video to obtain the process time of multiple processes includes: parsing the valid operation video to obtain the process time of multiple processes.

[0074] In this embodiment, validity processing is used to obtain valid operation videos from the operation videos, thereby eliminating content in the operation videos that is not related to the process. It should be understood that in the actual production process, the video captured by the camera may contain abnormal situations of the operators, such as trial production during production line debugging, irregular work behavior, etc. Therefore, the captured operation videos will contain some invalid videos. Therefore, in this embodiment, by performing validity processing on the operation videos, the invalid parts of the operation videos are eliminated, thereby obtaining valid videos.

[0075] Specifically, the validity processing can be performed by the user, obtaining a cut instruction input by the user, cutting the operation video, and obtaining a valid operation video. In this embodiment, the user watches the video, determines the starting point and end point of the valid video, and cuts the operation video.

[0076] In another embodiment, the validity processing is performed by an effective video learning model, which is a machine learning model. By inputting multiple sample homework videos for learning, an effective video learning model is trained. In this way, after obtaining the homework video, the homework video is input into the effective video learning model for learning, and the homework video can be processed for validity to obtain a valid homework video.

[0077] In one embodiment, the step of performing validity processing on the operation video to obtain a valid operation video includes: obtaining a preset allowance coefficient and a preset average working time; and performing cutting processing on the operation video based on the preset allowance coefficient and the preset average working time to obtain the valid operation video.

[0078] Specifically, the preset average working time is the average time it takes for multiple workers to complete a process, or it can be the average time it takes for the same worker to complete a process multiple times. The preset allowance factor is a factor that relaxes the preset average working time, so that the time length of the cropped effective work video falls within a certain range of the preset average working time. In this embodiment, the preset allowance factor (A) and the preset average working time (T) define the effective video time length (L) range: T(1-A)≤L≤T(1+A). Work videos that exceed the effective time length range are eliminated, and effective work videos are retained, thereby making the time length of the effective work video more accurate.

[0079] In one embodiment, the step of parsing the operation video to obtain the process time of multiple processes includes: segmenting the operation video in units of processes to obtain segmented videos of multiple processes; parsing each of the segmented videos to obtain the process time of multiple processes.

[0080] Specifically, a captured work video may contain multiple steps. Therefore, to obtain a video of a single step, the video needs to be segmented. This process is divided into multiple segments, each corresponding to a single step. Each segmented video is used to generate multiple segmented videos. By analyzing each segmented video, the process time for a single step can be determined.

[0081] In this embodiment, the segmentation process can be performed by the user, receiving a cropping instruction input by the user, and cropping the operation video into multiple segmented videos. In this embodiment, the user views the video, determines the multiple processes included in the operation video, and then crops the operation video by process, thereby obtaining multiple videos of a single process.

[0082] In another embodiment, the validity processing is performed by a segmented video learning model, which is a machine learning model. The segmented video learning model is trained by inputting multiple sample job videos and model segmented videos obtained after the model job video is cut and segmented to learn. In this way, after obtaining the job video, the job video is input into the segmented video learning model for learning, that is, the segmented video is output.

[0083] In one embodiment, the step of calculating the optimal clustering center of the process cluster based on the maximum expectation algorithm according to the normal distribution of the distance cluster also includes: obtaining the corrected process time calculated based on the newly collected operation video; and calculating the corrected optimal clustering center based on the process time and the corrected process time of n objects of the same process.

[0084] In this embodiment, after obtaining the optimal clustering center, videos are continuously collected, and the subsequently collected videos are newly collected operation videos. The newly collected operation videos are analyzed to obtain the process time of the newly collected operation videos. The process time of the newly collected operation videos is the corrected process time. The corrected process time is added to the process cluster to obtain an updated process cluster, and the cluster center of the updated process cumulative cluster is obtained, that is, the updated cluster center is obtained. The Euclidean distance between each process time and the corrected process time and the updated cluster center is calculated to obtain an updated distance cumulative cluster; according to the normal distribution of the updated distance cluster, the corrected optimal cluster center is calculated based on the maximum expectation algorithm.

[0085] In this embodiment, by continuously collecting operation videos and continuously inputting the process time of the operation videos into the process cluster for calculation, the optimal cluster center is continuously corrected, so that the optimal cluster center can more accurately reflect the standard working hours after correction.

[0086] In one embodiment, the method further includes: repeatedly executing the step of obtaining the corrected process time calculated based on the newly collected operation video, and calculating the corrected optimal clustering center based on the process time and the corrected process time of n objects of the same process, until the offset of the corrected optimal clustering center is less than the preset offset.

[0087] In this embodiment, the offset is the difference between the corrected optimal cluster center and the previously obtained optimal cluster center. In some embodiments, the offset is the difference between the corrected optimal cluster center and the original optimal cluster center. In this embodiment, after continuously correcting the optimal cluster center, when the difference between the corrected optimal cluster center and the original optimal cluster center is less than the preset offset, or when there is no further offset, cluster center C is now the standard working time for this process. This continuous correction makes the obtained standard working time more accurate.

[0088] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0089] Example 2

[0090] In this embodiment, please combine Figure 4 As shown, the specific implementation method:

[0091] 1. Use cameras installed at workstations on the production line to collect work videos of the same operator at this workstation in different time periods, as well as work videos of different operators, and upload the video information to the system, including the line name and job title of the job;

[0092] 2. Determine the validity of the collected work videos and define the effective video duration (L) range based on the job's allowance coefficient (A) and the average job working hours (T): T(1-A)≤L≤T(1+A). Eliminate work videos that exceed the effective duration range and retain valid work videos.

[0093] 3. Establish a standard working time determination model. The collected videos are segmented into process units. The same position contains an equal number of processes. Take one of the processes as an example. Assuming that a process is X and there are n objects in the valid position video, then the cluster X of this process is formed. , , ,..., }, assuming C is the cluster center of this type of cluster, the cluster is obtained by calculating the Euclidean distance from each object to the cluster center { , , ,..., }, according to the normal distribution of distance clusters combined with the maximum expectation algorithm, the optimal cluster center of this process cluster is obtained;

[0094] 4. As the collected data continues to increase, the new data will continue to be added to the process cluster. Repeat the above three steps, calculate in the standard working time determination model, and continuously optimize and correct the cluster center of this process cluster until the position of the cluster center converges and no significant displacement occurs. At this time, the cluster center C is the standard working time of this process.

[0095] 5. Input other processes into the standard working time determination model according to the above method and repeat the calculation to optimize the optimal cluster center for each process;

[0096] 6. Add up the standard working hours for all processes included in a position to obtain the standard working hours for this position, which can provide an evaluation standard for specific scenarios such as actual production or employee training.

[0097] It should be understood that the Expectation Maximum Algorithm (EXAM) is a machine learning-based optimization algorithm that uses iterative maximum likelihood estimation. It is generally used for parameter estimation in probabilistic models that include latent variables or missing data. Specifically, it uses existing sampled data to infer the most likely distribution parameters.

[0098] In this application, the distance from each object to the cluster center conforms to the normal distribution. If the Euclidean distance of each object is plotted into a histogram, the data in the middle of the histogram is the largest, the closer to the middle, the more data, and the farther away from the middle, the fewer data. When calculating, the largest data in the middle will be selected as representative, and the data farther away from the middle will be considered unrepresentative and will be screened out.

[0099] In this application, the distance clusters are first screened using the normal distribution, the specific selection range can be pre-set, and then the maximum expectation algorithm is used to determine the optimal cluster center.

[0100] The specific calculation method is as follows:

[0101] The input is the sample set X={ , , ,..., }, cluster center C, maximum number of iterations N.

[0102] 1) Randomly select k samples from the dataset as the initial centroid vectors { , , ,..., },

[0103] 2) n=1.2……N

[0104] First, the cluster partition X is initialized as =∅, t=1,2...k;

[0105] Secondly, for i=1,2...m, calculate the sample The distance from each centroid vector μj (j=1,2,...k): ,Will The category λi corresponding to dij has the smallest mark. At this time, update Cλi=Cλi∪{xi};

[0106] Again, for j=1,2,...,k, recalculate the new centroid for all sample points in Cj ;

[0107] Finally, if all k centroid vectors have not changed, go to step 3);

[0108] 3) Output cluster partition C={ , ,... }

[0109] Example 3

[0110] In this embodiment, Figure 2 As shown, a working time data calculation device is provided, comprising:

[0111] The acquisition module 210 is used to collect operation videos of preset workstations on the production line;

[0112] The process time obtaining module 220 is used to analyze the operation video and obtain the process time of multiple processes;

[0113] The process cluster forming module 230 is used to obtain n process times of the same process and form the process cluster;

[0114] A cluster center acquisition module 240 is used to obtain the cluster center of the process cluster;

[0115] The distance cluster acquisition module 250 is used to calculate the Euclidean distance between each process time and the cluster center to obtain a distance cluster;

[0116] The optimal cluster center obtaining module 260 is used to calculate the optimal cluster center of the process cluster based on the maximum expectation algorithm according to the normal distribution of the distance cluster.

[0117] In one embodiment, the acquisition module includes:

[0118] An operation video acquisition unit, used to acquire the operation video of a preset workstation on the production line;

[0119] A validity processing unit, configured to perform validity processing on the operation video to obtain a valid operation video;

[0120] The process time acquisition module is used to analyze the effective operation video to obtain the process time of multiple processes.

[0121] In one embodiment, the validity processing unit includes:

[0122] The coefficient and average working hours acquisition subunit is used to obtain the preset relaxation coefficient and the preset average working hours;

[0123] The cutting processing subunit is used to cut the operation video based on the preset relaxation coefficient and the preset average working hours to obtain the effective operation video.

[0124] In one embodiment, the process time acquisition module includes:

[0125] A segmentation processing unit, configured to segment the operation video in units of processes to obtain segmented videos of multiple processes;

[0126] The segmented parsing unit is used to parse each of the segmented videos to obtain the process time of multiple processes.

[0127] In one embodiment, the working time data calculation device further includes:

[0128] A correction process time calculation module is used to obtain the correction process time calculated based on the newly collected operation video;

[0129] The correction module is used to calculate the corrected optimal cluster center based on the process time and the corrected process time of n objects in the same process.

[0130] In one embodiment, the working time data calculation device also includes a repeated correction module, which is used to repeatedly execute the steps of obtaining the corrected process time calculated based on the newly collected work video, and calculating the corrected optimal clustering center based on the process time and the corrected process time of n objects of the same process, until the offset of the corrected optimal clustering center is less than the preset offset.

[0131] For the specific definition of the working time data calculation device, please refer to the definition of the working time data calculation method above, and will not be repeated here. Each unit in the above-mentioned working time data calculation device can be implemented in whole or in part by software, hardware, or a combination thereof. The above-mentioned units can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above-mentioned units.

[0132] Example 4

[0133] In this embodiment, a computer device is provided. Its internal structure diagram can be shown as follows: Figure 3As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program, and the non-volatile storage medium is deployed with a database, which stores a work video. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with other computer devices that have deployed application software. When the computer program is executed by the processor, a method for calculating working time data is implemented. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0134] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0135] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0136] Collect operation videos of preset workstations on the production line;

[0137] Analyze the operation video to obtain the process time of multiple processes;

[0138] Obtaining the process time of n objects of the same process to form the process cluster;

[0139] Obtaining the cluster center of the process cluster;

[0140] Calculating the Euclidean distance between each process time and the cluster center to obtain a distance accumulation cluster;

[0141] According to the normal distribution of the distance clusters, the optimal cluster center of the process clusters is calculated based on the maximum expectation algorithm.

[0142] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0143] Collecting the operation video of the preset workstation on the production line;

[0144] Performing validity processing on the operation video to obtain a valid operation video;

[0145] The effective operation video is analyzed to obtain the process time of multiple processes.

[0146] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0147] Get the preset slack factor and preset average working hours;

[0148] The operation video is cut based on the preset relaxation coefficient and the preset average working hours to obtain the effective operation video.

[0149] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0150] Segmenting the operation video based on the process to obtain segmented videos of multiple processes;

[0151] Analyze each of the segmented videos to obtain the process time of multiple processes.

[0152] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0153] Obtaining the correction process time calculated based on the newly collected operation video;

[0154] Based on the process time and the corrected process time of n objects in the same process, a corrected optimal cluster center is calculated.

[0155] In one embodiment, when the processor executes the computer program, the processor further implements the following steps:

[0156] Repeat the steps of obtaining the corrected process time calculated based on the newly collected operation video, and calculating the corrected optimal cluster center based on the process time and the corrected process time of n objects of the same process, until the offset of the corrected optimal cluster center is less than the preset offset.

[0157] Example 5

[0158] In this embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0159] Collect operation videos of preset workstations on the production line;

[0160] Analyze the operation video to obtain the process time of multiple processes;

[0161] Obtaining the process time of n objects of the same process to form the process cluster;

[0162] Obtaining the cluster center of the process cluster;

[0163] Calculating the Euclidean distance between each process time and the cluster center to obtain a distance accumulation cluster;

[0164] According to the normal distribution of the distance clusters, the optimal cluster center of the process clusters is calculated based on the maximum expectation algorithm.

[0165] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0166] Collecting the operation video of the preset workstation on the production line;

[0167] Performing validity processing on the operation video to obtain a valid operation video;

[0168] The effective operation video is analyzed to obtain the process time of multiple processes.

[0169] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0170] Get the preset slack factor and preset average working hours;

[0171] The operation video is cut based on the preset relaxation coefficient and the preset average working hours to obtain the effective operation video.

[0172] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0173] Segmenting the operation video based on the process to obtain segmented videos of multiple processes;

[0174] Analyze each of the segmented videos to obtain the process time of multiple processes.

[0175] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0176] Obtaining the correction process time calculated based on the newly collected operation video;

[0177] Based on the process time and the corrected process time of n objects in the same process, a corrected optimal cluster center is calculated.

[0178] In one embodiment, when the computer program is executed by a processor, the following steps are further implemented:

[0179] Repeat the steps of obtaining the corrected process time calculated based on the newly collected operation video, and calculating the corrected optimal cluster center based on the process time and the corrected process time of n objects of the same process, until the offset of the corrected optimal cluster center is less than the preset offset.

[0180] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0181] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0182] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for calculating working time data, characterized in that: include: Collect operation videos of preset workstations on the production line; Analyze the operation video to obtain the process time of multiple processes; Obtain the process time of n objects of the same process to form a process cluster; Obtaining the cluster center of the process cluster; Calculating the Euclidean distance between each process time and the cluster center to obtain a distance cluster; According to the normal distribution of the distance clusters, the optimal cluster center of the process clusters is calculated based on the maximum expectation algorithm, and the optimal cluster center is the required standard working hours; Obtaining the correction process time calculated based on the newly collected operation video; Calculating a corrected optimal cluster center based on the process time and the corrected process time of n objects in the same process; Repeating the steps of obtaining the corrected process time calculated based on the newly collected operation video, and calculating the corrected optimal cluster center based on the process time and the corrected process time of n objects in the same process, until the offset of the corrected optimal cluster center is less than a preset offset; The step of collecting the operation video of the preset workstation on the production line includes: Collecting the operation video of the preset workstation on the production line; Performing validity processing on the operation video to obtain a valid operation video; The step of analyzing the operation video to obtain the process time of multiple processes includes: Analyze the effective operation video to obtain the process time of multiple processes; The step of performing validity processing on the operation video to obtain a valid operation video includes: Get the preset slack factor and preset average working hours; Cutting the operation video based on the preset relaxation coefficient and the preset average working hours to obtain the effective operation video; Among them, the preset relaxation rate coefficient (A) and the preset average working hours (T) define the effective video duration (L) range: T(1-A)≤L≤T(1+A).

2. The method according to claim 1, characterized in that The step of analyzing the operation video to obtain the process time of multiple processes includes: Segmenting the operation video based on the process to obtain segmented videos of multiple processes; Analyze each of the segmented videos to obtain the process time of multiple processes.

3. A working time data calculation device, characterized in that: include: The acquisition module is used to collect operation videos of preset workstations on the production line; A process time acquisition module is used to analyze the operation video and obtain the process time of multiple processes; A process cluster forming module is used to obtain n process times of the same process to form a process cluster; A cluster center acquisition module, used to obtain the cluster center of the process cluster; A distance cluster obtaining module is used to calculate the Euclidean distance between each process time and the cluster center to obtain a distance cluster; An optimal cluster center obtaining module, configured to calculate the optimal cluster center of the process cluster based on the normal distribution of the distance cluster and the maximum expectation algorithm; A correction process time calculation module is used to obtain the correction process time calculated based on the newly collected operation video; A correction module, configured to calculate a corrected optimal cluster center based on the process time and the corrected process time of n objects in the same process; a repeated correction module, configured to repeatedly execute the steps of obtaining the corrected process time calculated based on the newly collected operation video, calculating the corrected optimal cluster center based on the process time and the corrected process time of n objects of the same process, until the offset of the corrected optimal cluster center is less than a preset offset; Wherein, the acquisition module includes: An operation video acquisition unit, used to acquire the operation video of a preset workstation on the production line; A validity processing unit, configured to perform validity processing on the operation video to obtain a valid operation video; The process time acquisition module is used to: Analyze the effective operation video to obtain the process time of multiple processes; The validity processing unit includes: The coefficient and average working hours acquisition subunit is used to obtain the preset relaxation coefficient and the preset average working hours; The cutting processing subunit is used to cut the operation video based on the preset relaxation coefficient and the preset average working hours to obtain the effective operation video.

4. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 2 are implemented.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 2 are implemented.

Images